Process for sorting plastic-containing material

Abstract

A process for sorting plastic-containing material, the process comprising providing a particulate plastic-containing material MS; subjecting the particulate material MS to sorting in a sorting unit US comprising a classification section SC for classifying particles of the particulate material MS with respect to their polyamide 6 content and polyamide 66 content, said classification section SC comprising an optical system configured for measuring at least two wavelengths and determining spectroscopic data of the particles of the particulate material MS therefrom and a processing unit UP configured for discriminating polyamide 6 comprised in a particle of MS from polyamide 66 comprised in said particle of MS by performing a computer-implemented classification method using said spectroscopic data; and further comprising a sorting section S S comprising a sorting device DS configured for sorting particles of M S classified in SC according to their respec- tive polyamide 6 and polyamide 66 content; wherein subjecting the particulate material MS to sorting comprises feeding the particulate material MS to the classification section SC; determin- ing spectroscopic data of the particles of MS with the optical system comprising measuring at least two wavelengths of a given particle in a wavelength range RW of from 0.9 to 12 µm; classi- fying the spectroscopically examined particles in the processing UP comprising performing said classification method using said spectroscopic data, obtaining particles classified according to their polyamide 6 and polyamide 66 content; and sorting the classified particles according to their respective polyamide 6 and polyamide 66 content in SS via the sorting device D S .

Description

Process for sorting plastic-containing materialThe present invention relates to a process for sorting plastic-containing material, to processes for recycling plastic-containing waste material, to a plastic-containing material sorting unit and to a classification method and a method for providing a trained model for discriminating different polyamides comprised in a plastic-containing material.There is an increasing need for recycling of plastic-containing material, such as engineering plastic material. Usually, many engineering plastic materials are black-colored. Often, said plastic materials are comminuted such as shreddered at their end of life, resulting in a particulate plastic-containing material, for example black-colored particles. For most of the promising recycling technologies including, but not restricted to mechanical recycling and depolymerization, said particulate material needs to be sorted with respect to their chemical composition.With known sorting systems, it may be possible to sort out different types of plastics, such as polyamides in general, especially black PA6 and black PA66, from a particulate mixed plasticcontaining material as a mixture of all polyamides.EP 3 954 473 A1 describes a method of processing waste for the recovery of various grades of plastics products for onward processing. The method includes receiving waste material; shredding the waste material; mechanically separating the waste material into at least three size fractions; removing at least some metallic objects from one of the at least three size fractions; splitting the one size fraction, after at least some metallic objects have been removed from it, into at least two streams; mechanically sorting the at least two streams, using near-infrared spectroscopy, to produce at least three product streams, a first product stream comprising substantially only one or more of polypropylene plastics, polyethylene plastics and polybromide plastics, a second product stream comprising substantially only one or more of polyethylene terephthalate (PET) plastics, polyethersulfone plastics and polycarbonate plastics, and a third product stream comprising substantially only one or more of polyurethane plastics, PVC plastics, polystyrene plastics, polyamide plastics, ABS, PTFE, latex and silicone plastics.E. H. Enlow et al. describe in “Discrimination of Nylon Polymers Using Attenuated Total Reflection Mid-infrared Spectra and Multivariate Statistical Techniques", Applied Spectroscopy, 2005;59(8):986-992, peak ratio analysis and multivariate statistics for the identification of nylon subclasses using attenuated total reflection (ATR) spectral data.M. N. Larsen et al. describe in Classification of black plastic types by hyperspectral imaging based on long-wave infrared emission spectroscopy , Polymer Testing, Volume 141 , 2024, 108629, a method where hyperspectral imaging in the long-wave infrared regime is used to distinguish between twelve samples of commercially available black plastics encompassing nine distinct polymer types.Despite the advantages achieved by known methods and devices, several technical challenges remain. Specifically, there is still a need for accurately separating PA6 polymer materials from PA66 polymer materials, in particular separating black-colored PA6 polymer materials from black-colored PA66 polymer materials, in particular separating shredded black-colored PA6 polymer materials from shredded black-colored PA66 polymer materials.It is therefore desirable to provide methods and devices which at least partially address the above-mentioned technical challenges. Specifically, methods and devices shall be proposed which address the need for separating PA6 polymer materials from PA66 polymer materials, in particular separating black-colored PA6 polymer materials from black-colored PA66 polymer materials, more particularly separating shredded black-colored PA6 polymer materials from shredded black-colored PA66 polymer materials.Therefore, the present invention relates to a process for sorting a plastic-containing material, the process comprising(i) providing a particulate plastic-containing material Ms;(ii) subjecting the particulate material Ms according to (i) to sorting in a sorting unit Us, said sorting unit Us comprising(a) a classification section Sc for classifying particles of the particulate material Ms with respect to their polyamide 6 content and polyamide 66 content, said classification section Sc comprising(a.1) an optical system configured for measuring at least two wavelengths and determining spectroscopic data of the particles of the particulate material Ms therefrom;(a.2) a processing unit UP configured for discriminating polyamide 6 comprised in a particle of Ms from polyamide 66 comprised in said particle of Ms by performing a computer-implemented classification method using the spectroscopic data according to (a.1);(b) a sorting section Ss comprising a sorting device Ds configured for sorting particles of Ms classified in Sc according to their respective polyamide 6 and polyamide 66 content;wherein subjecting the particulate material Ms to sorting comprises(11.1) feeding the particulate material Ms provided according to (i) to the classification section Sc according to (a);(11.2) determining spectroscopic data of x % of the particles of Ms with the optical system according to (a.1), comprising measuring at least two wavelengths of a given particle in a wavelength range Rw of from 0.9 to 12 pm, wherein x > 50;(11.3) classifying the particles spectroscopically examined according to (ii.2) in the processing UP according to (a.2), comprising performing the computer-implemented classification method using the spectroscopic data determined according to (ii.2), obtaining particles classified according to their polyamide 6 and polyamide 66 content;(11.4) sorting the particles classified according to (ii.3) according to their respective polyamide 6 and polyamide 66 content in Ss via the sorting device Ds, obtaining a sorted particulate plastic-containing material Ms.Preferably, the wavelength range Rw according to (ii.2) is of from 3 to 12 pm, more preferably of from 4 to 12 pm, preferably of from 5 to 12 pm, more preferably of from 6 to 12 pm, more preferably of from 7 to 12 pm, more preferably of from 8 to 12 pm.A preferred range of Rw according to (ii.2) is of from 0.9 to 2.5 pm. A further preferred range of Rw according to (ii.2) is of from 3 to 7 pm, more preferably of from 3 to 6 pm, more preferably of from 3 to 5 pm. A more preferred range of Rw according to (ii.2) is of from 6 to 12 pm, more preferably of from 7 to 12 pm, more preferably of from 8 to 12 pm.As far as the percentage of the particles of which spectroscopic data are determined according to (ii.2) is concerned, it is preferred that x is in the range of from 50 to 100, more preferably in the range of from 70 to 100, more preferably in the range of from 90 to 100, more preferably in the range of from 95 to 100. Conceivable ranges for x according to (ii.2) are from 96 to 100 or from 97 to 100 or from 98 to 100 or from 99 to 100. In particular, it is conceivable that x is 100.Regarding the particulate material Ms, it is preferred that it comprises, more preferably consists of a particulate plastic-containing waste material, more preferably of a particulate engineering plastic-containing waste material. Further preferably, the particulate plastic-containing material Ms comprises one or more of polyamide 6 and polyamide 66 and preferably further comprises carbon black. Preferably from 50 to 100 weight-%, more preferably from 70 to 100 weight-%, more preferably from 80 to 100 weight-%, more preferably from 90 to 100 weight-%, more preferably from 95 to 100 weight-% of the particulate plastic-containing material Ms consists of one or more of polyamide 6 and polyamide 66 and optionally or preferably additionally carbon black.In case less than 100 weight-% of the particulate material Ms consist of one or more of polyamide 6 and polyamide 66 and optionally or preferably carbon black, it is preferred that it is further comprising one or more components selected from the group consisting of at least one polymer compound other than polyamide 6 and polyamide 66, wood, glass, and at least one metal. The at least one polymer compound preferably comprises one or more of polypropylene, polycarbonate, polydimethylsiloxane, polyethylene, polyurethane, acrylonitrile-butadiene-styrene, acry- lonitrile-butadiene-styrene, acrylonitrile styrene acrylate, polyethylene terephthalate, polyester, polystyrene, polyvinylchloride, a copolymer of two or more thereof including statistical copolymers, gradient copolymers, alternating copolymers, block copolymers, and graft copolymers, polymer blends of two or more thereof, and a rubber material comprising one or more of a natural rubber material and a synthetic rubber material.Generally, there are no specific restrictions how the particulate material Ms is provided. Preferably according to the present invention, the particulate material Ms is a material derived from vehicles, preferably from end-of-life vehicles which are typically at least 15 years old. The vehicles can be passenger cars, light-duty or heavy-duty trucks, motorbikes, a utility vehicle, an agricultural vehicle, or recreational vehicles. The vehicles can be electric vehicles, such as a fully electric vehicle or a plug-in or mild hybrid electric vehicle. Preferably, providing the particulate material Ms comprises shredding a vehicle comprising polymeric vehicle parts, wherein from said shredding, an automotive shredder residue is obtained. Preferably, the automotive shredder residue is obtained by depollution of the vehicles, dismantling the vehicles, and shredding the vehicles. Preferably, the automotive shredder residue is obtainable by depollution of the vehicles, dismantling the vehicles, shredding the vehicles, and separating metal particles from the shredded vehicles. In depollution of vehicles, hazardous liquids such as fuel, lubricating oil, coolants, brake fluids and batteries can be removed from the vehicles prior to shredding. The dismantling of vehicles may comprise selective removal of parts, such as engines, gearboxes, tires, glass, and plastics, for being reused as spare parts for the second-hand market. The dismantling may also comprise the removal of larger plastic components, such as bumpers, dashboard, fluid containers for recycling the plastics separately. The automotive shredder residue may comprise further waste from other sources. For examples, garbage from the last owners may remain in the trunk or interior of the vehicles.The shredding can be made with a vehicle shredder machine. Vehicle shredder machines are manufactured in different sizes. Typically, a vehicle shredder machine comprises a heavy fastturning rotor, which may revolve in a vertical or a horizontal plane and is often equipped with swinging hammers. The vehicle shredder machine tears and shreds the car hulk until its partsare reduced to fragments with a desired fragment size, such as up to 30 cm, preferably 1 mm to 15 cm. Then the fragments may pass through grids and leave the rotor housing. The automotive shredder residue may represent about 10 to 40 weight-%, preferably from 15 to 35 weight-%, more preferably from 20 to 30 weight-% of the original vehicle weight.The automotive shredder residue may comprise fragments of various polymeric vehicle parts, such as fragments of bumpers, interior panels, dashboard, cable insulation, fuel tank, electrical insulation, flexible foam seating, foam insulation panels, automotive suspension bushings, electrical potting compounds, car body parts, pillar coverings, spoilers polymer parts coated with automotive paint, wheel covers, gears, bushes, cams, bearings, weatherproof coatings, interior and exterior trims, fuel systems, gear housings, headlamp retainer, engine cover, connector housings, door handles, carburetor components, exterior mirror components, windscreen wiper components, windscreen wiper protective housings, decorative grilles, cover strips, roof rails, window frames, sliding roof frames, antenna cladding covers, front and rear lights, radiator grill and body exterior parts, engine covers, cylinder head covers, intake pipes, cylinder head covers, engine covers, housings for charge air coolers, charge air cooler valves.The automotive shredder residue typically comprise fragments of various polymeric vehicle parts, such as fragments of bumpers, interior panels, dashboard, cable insulation, where these fragments are often made of polypropylene; fuel tank, electrical insulation, where these fragments are often made of polyethylene; flexible foam seating, foam insulation panels, automotive suspension bushings, electrical potting compounds, hard plastic parts, transmission mounts, motor mounts, seals, impact foam parts, where these fragments are often made of polyurethane; body parts, dashboards, wheel covers, where these fragments are often made of acryloni- trile-butadiene-styrene; gears, bushes, cams, bearings, charge air coolers, cylinder head covers, oil pans, engine cooling systems, thermostat and heater housings, exhaust systems including mufflers and housings for catalytic converters, air intake manifolds, timing chain belt front covers, where these fragments are often made of nylon 6 or nylon 6.6.; interior and exterior trims, fuel systems, small gears, where these fragments are often made of polyoxymethylene; wiper arm and gear housings, headlamp retainer, connector housings, where these fragments are often made of polyethylene terephthalate; and door handles, bumpers, carburetor components, where these fragments are often made of polybutylene terephthalate.The automotive shredder residue comprises preferably at least 30 weight-%, more preferably at least 40 weight-%, more preferably at least 50 weight-% of the fragments of the polymeric vehicle parts. Further, the automotive shredder residue preferably comprise at least 20 weight-%, more preferably at least 30 weight-%, more preferably at least 40 weight-% of the fragments of the polymeric vehicle parts, which are black polymeric vehicle parts. The black polymeric vehicle parts usually comprise carbon black pigments.Further, the automotive shredder residue may comprise up to 15 weight-%, preferably up to 10 weight-%, more preferably up to 5 weight-% of metal fragments, such as ferrous and non-ferrous metal particles. Still further, the automotive shredder residue may comprise up to 15 weight-%, preferably up to 10 weight-%, more preferably up to 5 weight-% of wood and cardboard. Yet further, the automotive shredder residue may comprise up to 15 weight-%, preferably up to 10 weight-%, more preferably up to 5 weight-% of glass fragments, e.g. broken window glass fragments.The automotive shredder residue can be separated into a shredder light fraction and a shredder heavy fraction. The separation of the light fraction and the heavy fraction can be achieved by air classification. Another air classification can be made by the rotary movement of the vehicle shredder machine may create a fanning action that can blow out the shredder light fraction, and the shredder heavy fraction may leave the vehicle shredder machine through a grid. The light fraction can be present in an amount of from 55 to 90 weight-%, preferably of from 65 to 85 weight-%, more preferably of from 70 to 80 weight-% of the automotive shredder residue. The heavy fraction may represent the remaining amount to 100 weight-%. The heavy fraction can be present in an amount of from 10 to 45 weight-%, preferably of from 15 to 35 weight-%, more preferably of from 20 to 30 weight-% of the automotive shredder residue. The light fraction may represent the remaining amount to 100 weight-%. Usually, the light fraction exhibits a lower weight percentage of rubber particles than the heavy fraction; a lower weight percentage of glass particles than the heavy fraction; a lower weight percentage of metal particles than the heavy fraction; a higher weight percentage of polyurethane foam particles than the heavy fraction; a lower weight percentage of solid and sand than the heavy fraction.Therefore, the present invention preferably relates to a process as defined herein, wherein providing the particulate material Ms according to (i) comprises(i.1) shredding a vehicle comprising polymeric vehicle parts, obtaining an automotive shredder residue (ASR);(i.2) preferably subjecting the automotive shredder residue according to (i.1) to one or more post-treatment steps (i.2.1) to (i.2.3):(i.2.1) separating metal fragments, including, but not restricted to one or more of ferrous metal fragments and non-ferrous metal fragments from the automotive shredder residue;(i.2.2.) separating the automotive shredder residue into a shredder light fraction and a shredder heavy fraction;(i.2.3) subjecting the automotive shredder residue, preferably one or more of the shredder light fraction and the shredder heavy fraction according to (i.2.2) to an aqueous treatment, said aqueous treatment preferably comprising one or more of washing and aqueous density separation; wherein the particulate material Ms is obtained from (i.1), preferably from (i.2).Further, the present invention preferably relates to a process as defined herein, wherein the particulate material Ms provided according to (i) is obtainable or obtained from a method comprising(i .1 ) shredding a vehicle comprising polymeric vehicle parts, obtaining an automotive shredder residue (ASR);(i.2) preferably subjecting the automotive shredder residue according to (i.1 ) to one or more post-treatment steps (i.2.1) to (i.2.3):(i.2.1) separating metal fragments, including, but not restricted to one or more of ferrous metal fragments and non-ferrous metal fragments from the automotive shredder residue;(i.2.2.) separating the automotive shredder residue into a shredder light fraction and a shredder heavy fraction;(i.2.3) subjecting the automotive shredder residue, preferably one or more of the the shredder light fraction and the shredder heavy fraction according to (i.2.2) to an aqueous treatment, said aqueous treatment preferably comprising one or more of washing and aqueous density separation; wherein the particulate material Ms is obtained from (i.1), preferably from (i.2).According to the present invention, a computer-implemented classification method is carried out according to (ii.3). Preferably, performing said computer-implemented classification method according to (ii.3) comprises(a) providing a computer-implemented trained classification model;(P) retrieving the trained model provided according to (a);(y) retrieving spectroscopic data of a particle of Ms determined according to (ii.2);(5) applying the trained model to the spectroscopic data according to (y), thereby discriminating polyamide 6 comprised from polyamide 66 and obtaining particles classified according to their polyamide 6 and polyamide 66 content.Preferably, providing the computer-implemented trained classification model according to (a) comprises(a.1) providing a computer-implemented trainable model;(a.2) training the trainable model provided according to (a.1) with a computer-implemented training method, the training method preferably comprising(a.2.1) retrieving a labelled spectroscopic training data set comprising spectroscopic data of polymer compounds determined by an optical system configured for measuring at least two wavelengths in the wavelength range Rw of polymer compounds having defined chemical composition;(a.2.2) training the trainable model on the labelled spectroscopic training data set, preferably by supervised training.According to (a.2.1), polymer compounds having defined chemical composition are employed. Generally, there are no specific restrictions regarding said chemical compositions with the proviso that said chemical compositions are suitable to train the trainable model with regard to the target polymers of the present invention, i.e. polyamide 6 and polyamide 66, and with regard to the conceivable particulate material Ms to be subjected to the sorting process of the present invention. Therefore, it is preferred that the polymer compounds having defined chemical composition include a polymer compound consisting of polyamide 6, a polymer compound consisting of polyamide 66, optionally include polymer compounds consisting of polyamide 6 and polyamide 66 optionally at different respective amounts, and further optionally include polymer compounds consisting of one or more of polyamide 6 and polyamide 66 and a further polymer component preferably selected from the group consisting of polypropylene, polycarbonate, polydimethylsiloxane, polyethylene, polyurethane, acrylonitrile-butadiene-styrene, acrylonitrile-butadi- ene-styrene, acrylonitrile styrene acrylate, polyethylene terephthalate, polyester, polystyrene, polyvinylchloride, a copolymer of two or more thereof including statistical copolymers, gradient copolymers, alternating copolymers, block copolymers, and graft copolymers, polymer blends of two or more thereof, a rubber material comprising one or more of a natural rubber material and a synthetic rubber material, and mixtures of two or more thereof.As far as the trainable model according to (a.1) is concerned, it is preferred that it comprises at least one model selected from the group consisting of a decision tree model, preferably at least one of an XGBoost model and a Random Forest model; a Support Vector Machine (SVM)model, preferably a linear kernel SVM; a nearest neighbors model, preferably a KNeighbors Classifier; a Bayes model, preferably a GaussianNB() model; a regression model, preferably at least one of a LogisticRegression, a linear and a nonlinear regression model including a regression model comprising transformed features which includes one or more of a log-transformed and polynomial model including a Partial Least Squares Discriminant Analysis (PLS-DA) model; an Artificial Neural Network (ANN), preferably a non-linear Artificial Neural Network (ANN), more preferably a deep learning architecture including one or more of Convolutional NN, Recurrent NN, a Long Short Term Memory NN; a kernel-based method; a tree regression model; a distributed gradient-boosting framework model, preferably a light gradient-boosting machine model (LightGBM classifier); a soft independent modelling by class analogy (SIMCA); a Multi Curve Resolution (MCR) model; a t-distributed stochastic neighbor embedding (t-SNE) model; a Uniform Manifold Approximation and Projection for Dimension Reduction (UMAP) model; a Classical Least Squares (CLS) model; a Simultaneous Component Analysis (ASCA-ANOVA) model; and a Multi-Level Simultaneous Component Analysis (MLSCA) model.Preferably, the spectroscopic data according to (a.2.1) are determined with an optical system configured for measuring two or more wavelengths, preferably three or more wavelengths, more preferred five or more wavelengths, in particular a hyperspectral camera CH having a spectral resolution of 150 to 200 spectral bands, preferably in the wavelength range Rw as defined herein.According to (a.2), the training method preferably further comprises pre-processing the spectroscopic data determined by CH to obtain the spectroscopic training data set. Said pre-processing preferably comprises applying a Standard Normal Variate (SNV) to the spectroscopic data obtained. The pre-processing further preferably comprises taking at least one derivative of the spectroscopic data determined by CH, more preferably of the SNV pre-processed spectroscopic data. Yet further preferably, the pre-processing further comprises normalizing the spectroscopic data determined by CH, more preferably normalizing the at least one derivative of the spectroscopic data, more preferably normalizing the at least one derivative of the SNV pre-processed spectroscopic data.According to (a.2), the training method preferably still further comprises at least one cross validation in the spectroscopic training data set to obtain at least one pair of training data set and test data set.According to (a.2), the training method preferably yet further comprises at least one validation step comprising testing the accuracy of the training by using the test data set.According to (5), the trained model is applied to the spectroscopic data according to (y). Preferably, applying the trained model according to (5) comprises attributing a classification probability to a given particle of Ms, wherein the particle is classified as a polyamide 6 particle if the polyamide 6 classification probability is at least 0.6, preferably at least 0.7, more preferably at least 0.8, more preferably at least 0.9, and wherein the particle is classified as a polyamide 66 particle if the polyamide 66 classification probability is at least 0.6, preferably at least 0.7, more preferably at least 0.8, more preferably at least 0.9.According to the present invention, a preferred trained model is a trained Partial Least Squares Discriminant Analysis (PLS-DA) model. In this case, it is preferred that the classification probability of the given particle of Ms is determined by using classification results obtained by applying the PLS-DA model to the spectroscopic data of the particle of Ms. Specifically, the classification probability may be determined using a regression vector of the PLS-DA model applied to the spectroscopic data of the given particle of Ms.As far as the design of the sorting device Ds according to the present invention is concerned, there are generally no specific restrictions, with the proviso that the classified particles can be suitably separated.Preferably, the sorting device Ds comprised in the sorting section Ss comprises three outlet devices DOPA6, DOPA66 and DON, and wherein sorting the classified particles according to (ii.4) comprises controlling the sorting device Ds so that a particle classified as a polyamide 6 particle is passed to DOPA6, a particle classified as a polyamide 66 particle is passed to DOPA66, and a particle classified neither as a polyamide 6 particle nor as a polyamide 66 particle is passed to DON.More preferably, the sorting device Ds comprises a first sorting subdevice Dssi, preferably a first pneumatic sorting subdevice or a first mechanical sorting subdevice, configured for actively removing a particle classified as a polyamide 6 particle to DOPA6 from the particulate material Ms or for actively removing a particle classified as a polyamide 66 particle to DOPA66 from the particulate material Ms; and a second sorting subdevice Dss2, preferably a second pneumatic sorting subdevice or a second mechanical sorting subdevice, configured for actively removing a particle classified as a polyamide 66 particle to DOPA66 or for actively removing a particle classified as a polyamide 6 particle to DOPA6 from the remainder material.Therefore, in case the particulate material Ms provided according to (i) comprises particles classified as polyamide 6 particles according to (ii.3), particles classified as polyamide 66 particles according to (ii.3), and optionally particles classified neither as polyamide 6 particles nor as polyamide 66 particles, the process preferably comprises(11.4.1) actively removing particles classified as polyamide 6 particles to DOPA6 from the particulate material Ms via Dssi, obtaining sorted-out particles classified as polyamide 6 particles and further obtaining a first remainder material MSRI comprising particles classified as polyamide 66 particles and optionally particles neither classified as polyamide 66 particles nor classified as polyamide 6 particles;(11.4.2) actively removing particles classified as polyamide 66 particles to DOPA66 from the remained material according to (ii.4.1) via Dss2, obtaining sorted-out particles classified as polyamide 66 particles and optionally further obtaining a second remainder material MSR2 comprising particles classified neither as polyamide 66 particles nor as polyamide 6 particles, said second remainder stream being passed to DON; or(ii.4.1’) actively removing particles classified as polyamide 66 particles to DOPA66 from the particulate material Ms via Dssi, obtaining sorted-out particles classified as polyamide 66 particles and further obtaining a first remainder material MSRI comprising particles classified as polyamide 6 particles and optionally particles neither classified as polyamide 66 particles nor classified as polyamide 6 particles;(ii .4.2’) actively removing particles classified as polyamide 6 particles to DOPA6 from the remained material according to (ii.4.1) via Dss2, obtaining sorted-out particles classified as polyamide 6 particles and optionally further obtaining a second remainder material MSR2 comprising particles classified neither as polyamide 6 particles nor as polyamide 66 particles, said second remainder stream being passed to DON.Preferably, controlling Ds, and preferably Dssi and Dss2, is computer-controlled.From stage (ii) of the process of the present invention, a sorted particulate plastic material is obtained. Depending on the chemical composition of the particles having been subjected to the sorting process, one or more of a particulate material of the particles classified as polyamide 6 particles, a particulate material of the particles classified as polyamide 66 particles, and a particulate material of the particles classified neither as polyamide 6 particles nor polyamide 66 particles is / are obtained. In a particularly preferred subsequent downstream stage (iii), the particulate material of the particles classified as polyamide 66 particles and / or the particulate material of the particles classified as polyamide 66 particles is / are passed to one or more suitable recycling steps.Therefore, the present invention also relates to a process as described herein, further comprising(iii) passing sorted plastic-containing material to recycling, said recycling comprising one or more of (iii.1) and (iii.2):(111.1) passing sorted plastic-containing material classified as polyamide 6 plastic material, preferably sorted-out particles classified as polyamide 6 particles obtained according to (ii.4.1) or (ii.4.2’) as described herein to one or more of depolymerization, including one or more of neutral, acidic and alkaline depolymerization, and enzymatic depolymerization; solvolysis, preferably to solvolysis with one or more alcohols, more preferably with methanol; solvent-based recycling;(111.2) passing sorted plastic-containing material classified as polyamide 66 plastic material, preferably sorted-out particles classified as polyamide 66 particles obtained according to (ii.4.2) or (ii.4.1’) as described herein to one or more of depolymerization, including one or more of neutral, acidic and alkaline depolymerization, and enzymatic depolymerization; solvolysis, preferably to solvolysis with one or more alcohols, more preferably with methanol; solvent-based recycling.Consequently, in view of the preferred downstream recycling stage (iii), the present invention also relates to a process for recycling plastic-containing waste material, comprising sorting plastic-containing waste material according to a process as described herein and comprising the stages (i) and (ii), the recycling process further comprising(iii) passing sorted plastic-containing material to recycling, said recycling comprising one or more of (iii.1) and (iii.2):(iii.1) passing sorted plastic-containing material classified as polyamide 6 plastic material, preferably sorted-out particles classified as polyamide 6 particles obtained according to (ii.4.1) or (ii.4.2’) as described herein to one or more of depolymerization, including one or more of neutral, acidic and alkaline depolymerization, and enzymatic depolymerization; solvolysis, preferably to solvolysis with one or more alcohols, more preferably with methanol; solvent-based recycling;(iii.2) passing sorted plastic-containing material classified as polyamide 66 plastic material, preferably sorted-out particles classified as polyamide 66 particles obtained according to (ii.4.2) or (ii.4.1 ’) as described herein to one or more of depolymerization, including one or more of neutral, acidic and alkaline depolymerization, and enzymatic depolymerization; solvolysis, preferably to solvolysis with one or more alcohols, more preferably with methanol; solvent-based recycling.Still further, the present invention relates to a process for recycling plastic-containing waste material, comprising(iii) passing sorted plastic-containing material to recycling, said sorted plastic-containing material being obtainable or obtained by a a process as described herein and comprising the stages (i) and (ii); said recycling comprising one or more of (iii.1) and (iii.2):(111.1) passing sorted plastic material classified as polyamide 6 plastic material, preferably sorted-out particles classified as polyamide 6 particles obtained according to (ii.4.1 ) or (ii.4.2’) as described herein to one or more of depolymerization, including one or more of neutral, acidic and alkaline depolymerization, and enzymatic depolymerization; solvolysis, preferably to solvolysis with one or more alcohols, more preferably with methanol; solvent-based recycling;(111.2) passing sorted plastic-containing material classified as polyamide 66 plastic material, preferably sorted-out particles classified as polyamide 66 particles obtained according to (ii.4.2) or (ii.4.1 ’) as described herein to one or more of depolymerization, including one or more of neutral, acidic and alkaline depolymerization, and enzymatic depolymerization; solvolysis, preferably to solvolysis with one or more alcohols, more preferably with methanol; solvent-based recycling.In addition to the processes as described, the present invention further relates to a sorting unit Us for sorting a plastic-containing material Ms, preferably for carrying out a sorting process as described herein and comprising the stages (i) and (ii), wherein the sorting unit Us comprises (a) a classification section Sc for classifying particles of the particulate material Ms with respect to their polyamide 6 content and polyamide 66 content, said classification section Sc comprising(a.1) an optical system configured for determining spectroscopic data of the particles of the sorting material by measuring at least two wavelengths of a given particle of Ms in a wavelength range Rw of from 0.9 to 12 pm, preferably of from 3 to 12 pm, more preferably of from 4 to 12 pm, preferably of from 5 to 12 pm, more preferably of from 6 to 12 pm, more preferably of from 7 to 12 pm, more preferably of from 8 to 12 pm;(a.2) a processing unit UP configured for discriminating polyamide 6 comprised a particle of Ms from polyamide 66 comprised in said particle of Ms by performing a computer- implemented classification method using the spectroscopic data according to (a.1);(b) a sorting section Ss comprising a sorting device Ds configured for sorting particles of Ms classified in Sc according to their respective polyamide 6 and polyamide 66 content, the sorting section Ss preferably further comprising three outlet devices DOPA6, DOPA66 and DON.Preferably, the sorting device Ds which is preferably computer-controlled comprises a first sorting subdevice Dssi, preferably a first pneumatic sorting subdevice or a first mechanical sorting subdevice, configured for actively removing a particle classified as a polyamide 6 particle to DOPA6 from the particulate material Ms or for actively removing a particle classified as a polyamide 66 particle to DOPA66 from the particulate material Ms; and a second sorting subdevice Dss2, preferably a second pneumatic sorting subdevice or a second mechanical sorting subdevice, configured for actively removing a particle classified as a polyamide 66 particle to DOPA66 or for actively removing a particle classified as a polyamide 6 particle to DOPA6 from the remainder material.classification methodStill further, the present invention relates to a computer-implemented classification method for discriminating polyamide 6 from polyamide 66, preferably for discriminating polyamide 6 comprised in a particle of a particulate plastic-containing material Ms from polyamide 66 comprised in said particle of Ms, said computer-implemented classification method comprising (a) providing a computer-implemented trained classification model;(P) retrieving the trained model provided according to (a);(y) retrieving spectroscopic data of polymer compounds comprising one or more of polyamide 6 and polyamide 66, said spectroscopic data being determined by an optical system configured for measuring at least two wavelengths in a wavelength range Rw of from 0.9 to 12 pm;(5) applying the trained model to the spectroscopic data according to (y), thereby discriminating polyamide 6 from polyamide 66;wherein Rw according to (y) is preferably of from 3 to 12 pm, more preferably of from 4 to 12 pm, more preferably of from 5 to 12 pm, more preferably of from 6 to 12 pm, more preferably of from 7 to 12 pm, more preferably of from 8 to 12 pm.With regard to the trained model which is provided according to (a), it is preferred that providing the trained model according to (a) comprises using at least one computer-implemented training method, wherein (a) further comprises(a.1) providing a computer-implemented trainable model;(a.2) training the trainable model provided according to (a.1) with a computer-implemented training method, the training method comprising(a.2.1) retrieving a labelled spectroscopic training data set, the spectroscopic training data set comprising spectroscopic data of polymer compounds determined by an optical system in the wavelength range Rw of polymer compounds having defined chemical composition;(a.2.2) training the trainable model on the labelled spectroscopic training data set, preferably by supervised training.With regard to details concerning the individual steps (a) to (5), general reference is made to the disclosure herein.Specifically according to (a.2.1), polymer compounds having defined chemical composition are employed. Generally, there are no specific restrictions regarding said chemical compositions with the proviso that said chemical compositions are suitable to train the trainable model with regard to the target polymers of the present invention, i.e. polyamide 6 and polyamide 66, and with regard to the conceivable particulate material MS to be subjected to the sorting process of the present invention. Therefore, it is preferred that the polymer compounds having defined chemical composition include a polymer compound consisting of polyamide 6, a polymer compound consisting of polyamide 66, optionally include polymer compounds consisting of polyamide 6 and polyamide 66 optionally at different respective amounts, and further optionally include polymer compounds consisting of one or more of polyamide 6 and polyamide 66 and a further polymer component preferably selected from the group consisting of polypropylene, polycarbonate, polydimethylsiloxane, polyethylene, polyurethane, acrylonitrile-butadiene-styrene, acry- lonitrile-butadiene-styrene, acrylonitrile styrene acrylate, polyethylene terephthalate, polyester, polystyrene, polyvinylchloride, a copolymer of two or more thereof including statistical copolymers, gradient copolymers, alternating copolymers, block copolymers, and graft copolymers, polymer blends of two or more thereof, a rubber material comprising one or more of a natural rubber material and a synthetic rubber material, and mixtures of two or more thereof.Further, as far as the trainable model according to (a.1) is concerned, it is preferred that it comprises at least one model selected from the group consisting of a decision tree model, preferably at least one of an XGBoost model and a Random Forest model; a Support Vector Machine (SVM) model, preferably a linear kernel SVM; a nearest neighbors model, preferably a KNeigh- bors Classifier; a Bayes model, preferably a GaussianNB() model; a regression model, preferably at least one of a LogisticRegression, a linear and a nonlinear regression model including a regression model comprising transformed features which includes one or more of a log-transformed and polynomial model including a Partial Least Squares Discriminant Analysis (PLS-DA) model; an Artificial Neural Network (ANN), preferably a non-linear Artificial Neural Network (ANN), more preferably a deep learning architecture including one or more of Convolutional NN, Recurrent NN, a Long Short Term Memory NN; a kernel-based method; a tree regression model; a distributed gradient-boosting framework model, preferably a light gradient-boosting machine model (LightGBM classifier); a soft independent modelling by class analogy (SIMCA); a Multi Curve Resolution (MCR) model; a t-distributed stochastic neighbor embedding (t-SNE) model; a Uniform Manifold Approximation and Projection for Dimension Reduction (UMAP) model; a Classical Least Squares (CLS) model; a Simultaneous Component Analysis (ASCA- ANO A) model; and a Multi-Level Simultaneous Component Analysis (MLSCA) model.Preferably, the spectroscopic data according to (a.2.1) are determined with an optical system configured for measuring two or more wavelengths, preferably three or more wavelengths, more preferred five or more wavelengths, in particular a hyperspectral camera CH having a spectral resolution of 150 to 200 spectral bands, preferably in the wavelength range Rw as defined herein.According to (a.2), the training method preferably further comprises pre-processing the spectroscopic data determined by CH to obtain the spectroscopic training data set. Said pre-processing preferably comprises applying a Standard Normal Variate (SNV) to the spectroscopic data obtained. The pre-processing further preferably comprises taking at least one derivative of the spectroscopic data determined by CH, more preferably of the SNV pre-processed spectroscopic data. Yet further preferably, the pre-processing further comprises normalizing the spectroscopic data determined by CH, more preferably normalizing the at least one derivative of the spectroscopic data, more preferably normalizing the at least one derivative of the SNV pre-processed spectroscopic data.Further according to (a.2), the training method preferably still further comprises at least one cross validation in the spectroscopic training data set to obtain at least one pair of training data set and test data set.Still further according to (a.2), the training method preferably yet further comprises at least one validation step comprising testing the accuracy of the training by using the test data set.According to (5), the trained model is applied to the spectroscopic data according to (y). Preferably, applying the trained model according to (5) comprises attributing a classification probability to a given particle of Ms, wherein the particle is classified as a polyamide 6 particle if the polyamide 6 classification probability is at least 0.6, preferably at least 0.7, more preferably at least 0.8, more preferably at least 0.9, and wherein the particle is classified as a polyamide 66 particle if the polyamide 66 classification probability is at least 0.6, preferably at least 0.7, more preferably at least 0.8, more preferably at least 0.9.According to the present invention, a preferred trained model is a trained Partial Least Squares Discriminant Analysis (PLS-DA) model. In this case, it is preferred that the classification probability of the given particle of Ms is determined by using classification results obtained by applying the PLS-DA model to the spectroscopic data of the particle of Ms. Specifically, the classification probability may be determined using a regression vector of the PLS-DA model applied to the spectroscopic data of the given particle of Ms.According to the present invention, it is particularly preferred that the classification method as described above is carried out in step (ii.3) of a sorting process as described herein.Trained / training modelStill further, the present invention relates to a method for providing a trained model for discriminating polyamide 6 from polyamide 66, preferably for discriminating polyamide 6 comprised a particle of a particulate plastic-containing material Ms from polyamide 66 comprised in said particle of Ms, said method comprising(a.1) providing a computer-implemented trainable model;(a.2) training the trainable model provided according to (a.1) with a computer-implemented training method, the training method comprising(a.2.1) retrieving a labelled spectroscopic training data set, the spectroscopic training data set comprising spectroscopic data of polymer compounds determined by anoptical system configured for measuring at least two wavelengths in a wavelength range Rw of from 0.9 to 12 pm of polymer compounds having defined chemical composition;(a.2.2) training the trainable model on the labelled spectroscopic training data set, preferably by supervised training; wherein Rw according to (a.2.1) is preferably of from 3 to 12 pm, more preferably of from 4 to 12 pm, more preferably of from 5 to 12 pm, more preferably of from 6 to 12 pm, more preferably of from 7 to 12 pm, more preferably of from 8 to 12 pm.With regard to details concerning the individual steps (a.1) and (a.1), general reference is made to the disclosure herein.Specifically according to (a.2.1), polymer compounds having defined chemical composition are employed. Generally, there are no specific restrictions regarding said chemical compositions with the proviso that said chemical compositions are suitable to train the trainable model with regard to the target polymers of the present invention, i.e. polyamide 6 and polyamide 66, and with regard to the conceivable particulate material MS to be subjected to the sorting process of the present invention. Therefore, it is preferred that the polymer compounds having defined chemical composition include a polymer compound consisting of polyamide 6, a polymer compound consisting of polyamide 66, optionally include polymer compounds consisting of polyamide 6 and polyamide 66 optionally at different respective amounts, and further optionally include polymer compounds consisting of one or more of polyamide 6 and polyamide 66 and a further polymer component preferably selected from the group consisting of polypropylene, polycarbonate, polydimethylsiloxane, polyethylene, polyurethane, acrylonitrile-butadiene-styrene, acry- lonitrile-butadiene-styrene, acrylonitrile styrene acrylate, polyethylene terephthalate, polyester, polystyrene, polyvinylchloride, a copolymer of two or more thereof including statistical copolymers, gradient copolymers, alternating copolymers, block copolymers, and graft copolymers, polymer blends of two or more thereof, a rubber material comprising one or more of a natural rubber material and a synthetic rubber material, and mixtures of two or more thereof.Further, as far as the trainable model according to (a.1) is concerned, it is preferred that it comprises at least one model selected from the group consisting of a decision tree model, preferably at least one of an XGBoost model and a Random Forest model; a Support Vector Machine (SVM) model, preferably a linear kernel SVM; a nearest neighbors model, preferably a KNeigh- bors Classifier; a Bayes model, preferably a GaussianNB() model; a regression model, preferably at least one of a LogisticRegression, a linear and a nonlinear regression model including aregression model comprising transformed features which includes one or more of a log-transformed and polynomial model including a Partial Least Squares Discriminant Analysis (PLS-DA) model; an Artificial Neural Network (ANN), preferably a non-linear Artificial Neural Network (ANN), more preferably a deep learning architecture including one or more of Convolutional NN, Recurrent NN, a Long Short Term Memory NN; a kernel-based method; a tree regression model; a distributed gradient-boosting framework model, preferably a light gradient-boosting machine model (LightGBM classifier); a soft independent modelling by class analogy (SIMCA); a Multi Curve Resolution (MCR) model; a t-distributed stochastic neighbor embedding (t-SNE) model; a Uniform Manifold Approximation and Projection for Dimension Reduction (UMAP) model; a Classical Least Squares (CLS) model; a Simultaneous Component Analysis (ASCA- ANOVA) model; and a Multi-Level Simultaneous Component Analysis (MLSCA) model.Preferably, the spectroscopic data according to (a.2.1) are determined with an optical system configured for measuring two or more wavelengths, preferably three or more wavelengths, more preferred five or more wavelengths, in particular a hyperspectral camera CH having a spectral resolution of 150 to 200 spectral bands, preferably in the wavelength range Rw as defined herein.According to (a.2), the training method preferably further comprises pre-processing the spectroscopic data determined by CH to obtain the spectroscopic training data set. Said pre-processing preferably comprises applying a Standard Normal Variate (SNV) to the spectroscopic data obtained. The pre-processing further preferably comprises taking at least one derivative of the spectroscopic data determined by CH, more preferably of the SNV pre-processed spectroscopic data. Yet further preferably, the pre-processing further comprises normalizing the spectroscopic data determined by CH, more preferably normalizing the at least one derivative of the spectroscopic data, more preferably normalizing the at least one derivative of the SNV pre-processed spectroscopic data.Further according to (a.2), the training method preferably still further comprises at least one cross validation in the spectroscopic training data set to obtain at least one pair of training data set and test data set.Still further according to (a.2), the training method preferably yet further comprises at least one validation step comprising testing the accuracy of the training by using the test data set.Further, the present invention relates to a trained model for discriminating polyamide 6 from polyamide 66, provided by a method as described above.According to the present invention, it is particularly preferred that the training method as described above is carried out in step (a) of a computer-implemented classification method as described herein which in turn is particularly preferably carried out in step (ii.3) of a sorting process as described herein.Computer program and storage mediumThe present invention also relates to a computer program, comprising instructions which, when the program is executed by at least one processor of a computer or a computer network, cause the processor to perform the training method as described herein.Further, the present invention relates to a computer program, comprising instructions which, when the program is executed by at least one processor of a computer or a computer network, cause the processor to perform the classification method as described herein.Yet further, the present invention relates to a computer-readable storage medium, preferably a non-transient computer-readable storage medium, comprising instructions which, when the instructions are executed by at least one processor of a computer or a computer network, cause the processor to perform the training method as described herein.Still further, the present invention relates to computer-readable storage medium, preferably a non-transient computer-readable storage medium, comprising instructions which, when the instructions are executed by at least one processor of a computer or a computer network, cause the processor to perform the classification method as described herein.Further aspectsThe present invention also relates to a sorted plastic-containing material, preferably a sorted plastic-containing waste material, obtainable or obtained by a sorting process as described herein, comprising, preferably consisting of, the stages (i) and (ii) as described herein.Further, the present invention relates to a method for recycling a plastic-containing material, comprising providing a sorted plastic-containing material, preferably a sorted plastic-containing waste material, said sorted material being obtainable or obtained by a sorting process as described herein, comprising, preferably consisting of, the stages (i) and (ii) as described herein,and / or by using the sorting unit Us as described herein, and recycling said sorted plastic-containing material to obtain a polymer; wherein the method for recycling preferably comprises, more preferably consists of one or more of mechanical recycling, chemical recycling and sol- vent-based recycling.Further, the present invention relates to the use of a sorted plastic-containing material, preferably a sorted plastic-containing waste material, in a recycling process, wherein the sorted material is obtainable or obtained by a process as described herein, comprising, preferably consisting of, the stages (i) and (ii) as described herein and / or by using the sorting unit Us as described herein.According to another aspect, the present invention relates to a process, preferably a sorting process as defined herein, comprising the step of converting a chemical material obtainable or obtained by said process to obtain a product Q.According to yet another aspect, the present invention relates to a process, comprising the step of using the unit Us as defined herein to obtain a chemical material; and preferably converting the chemical material to obtain a product Q.Preferably, the product Q is selected from: building block or monomer; or polymer, preferably polymer A, polymer composition, preferably polymer composition A, or polymer product, preferably polymer product A; or industrial use polymer, industrial use surfactant, descaling compound, industrial use biocide, industrial use solvent, industrial use dispersant, composition thereof or formulation thereof; or agrochemical composition, agrochemical formulation auxiliary or agrochemically active ingredient; or active pharmaceutical ingredient or intermediate thereof, pharmaceutical excipient, animal feed additive, human food additive, dietary supplements, aroma chemical or aroma composition; or aqueous polymer dispersion, preferably polyurethane or polyurethane - poly(meth)acry- late hybrid polymer dispersion, emulsion, binder for paper and fiber coatings, UV-curable acrylic polymer for hot melts and coatings polyisocyanates, hyperbranched polyester polyol, polymeric dispersant for inorganic binder compositions, unsaturated polyester polyol or 100% curable composition; orcosmetic surfactant, emollient, wax, cosmetic polymer, UV filter, further cosmetic ingredient or composition or formulation thereof; or polymer B, polymer composition B, coating composition, other functional composition, foil, molded body, coating or coated substrate.Preferably, the content of the chemical material in the product Q is 1 weight-% or more, preferably 2 weight-% or more, more preferably 5 weight-% or more, more preferably 15 weight-% or more, more preferably 30 weight-% or more, more preferably 40 weight-% or more, more preferably 60 weight-% or more, more preferably 80 weight-% or more, more preferably 90 weight-% or more, more preferably 95 weight-% or more; and / or the content of the chemical material in the product Q is 100 weight-% or less, preferably 95 weight-% or less, more preferably 90 weight-% or less, more preferably 50 weight-% or less, more preferably 25 weight-% or less, more preferably 10 weight-% or less; and preferably wherein the content is determined based on identity preservation and / or segregation and / or mass balance and / or book and claim chain of custody models, preferably based on mass balance, preferably the International Sustainability and Carbon Certification (ISCC) standard.The publication Prior Art Disclosure; Issue 684; paragraphs

[1000] to

[8005] ; ISSN: 2198-4786; published: February 12, 2024 will be regarded as Reference RF1 , which is incorporated herein by reference in its entirety. Preferably, the product Q referred to in the preceding paragraph is a product as described in Reference RF1 ; paragraphs

[1000] to

[8005] , Preferably, the process described herein is further a process for the production of a product, preferably product Q.The converting step to obtain the product Q preferably comprises one or more step(s) as described below and can be performed by conventional methods well known to a person skilled in the art. The converting step preferably comprises one or more step(s) selected from: recycling, preferably depolymerizing, gasifying, pyrolyzing, and / or steam cracking; and / or purifying, preferably crystallizing, (solvent) extracting, distilling, evaporating, hydrotreating, absorbing, adsorbing and / or subjecting to ion exchanger; and / or assembling, preferably foaming, synthesizing, chemical conversion, chemically transforming, polymerizing and / or compounding; and / or forming, preferably foaming, extruding and / or molding; and / or finishing, preferably coating and / or smoothing.In addition, the one or more step(s) are described in detail in Reference RF1 ; paragraphs

[1000] to

[8005] ,The term “building block”, as used in the context of the product Q herein, comprises compounds, which are in a gaseous or liquid state under standard conditions of 0°C and 0.1 MPa. Building blocks are typically used in chemical industry to form secondary products, which provide a high-er structural complexity and / or higher molecular weight than the building block on which the sec-ondary product is based. The building block is preferably selected from the group consisting of hydrogen, carbon monoxide, carbon dioxide, ethylene oxide, ethylene glycols, syngas compris-ing a mixture of hydrogen and carbon monoxide, alkanes, alkenes, alkynes and aromatic com-pounds. The alkanes, alkenes, alkynes and aromatic compounds comprise in particular 1 to 12 carbon atoms, respectively.The term “monomer”, as used in the context of the product Q herein, comprises molecules, which can react with each other to form polymer chains by polymerization. The monomer is preferably selected from the group consisting of (meth)acrylic acid, salts of (meth)acrylic acid; in particular sodium, potassium and zinc salts; (meth)acrolein and (meth)acrylates. (Methacrylates comprising 1 to 22 carbon atoms are preferred, in particular comprising 1 to 8 carbon atoms. The terms (meth)acrylic acid, (meth)acrolein or (meth)acrylate relate to acrylic acid, acrolein or acrylate and also to methacrylic acid, methacrolein or methacrylate, where applicable. Further, the monomer can be selected from hexamethylenediamine (HMD) and adipic acid.The building block can further be an intermediate compound. The term “intermediate compound”, as used in the context of the product Q herein, comprises organic reagents, which are applied for formation of compounds with higher molecular complexity. The intermediate compound can be selected for example from the group consisting of phosgene, polyisocyanates and propylene oxide. The polyisocyanates are in particular aromatic di- and polyisocyanates, preferably toluene diisocyanate (TDI) and / or diphenylmethane diisocyanate (MDI).The building block and the monomer and typical converting step(s) to obtain the building block or monomer are described in more detail in paragraphs

[1000] to

[1012] of Reference RF1.The term “polymer A”, as used in the context of the product Q herein, comprises thermoplastic, e.g., polyamide or thermoplastic polyurethane, thermoset, e.g., polyurethane, elastomer, e.g., polybutadiene, or a copolymer or a mixture thereof and is defined in more detail in paragraphs

[2001] to

[2007] of Reference RF1.The term “polymer composition A”, as used in the context of the product Q herein, comprises all compositions comprising a polymer as described above and one or more additive(s), e.g. rein-forcement, colorant, modifier and / or flame retardant, and is defined in more detail in paragraph

[2008] of Reference RF1.The term “polymer product A”, as used in the context of the product Q herein, comprises any product comprising the polymer A and / or polymer composition A as described above and is defined in more detail in paragraphs

[2009] and

[2010] of Reference RF1.The step(s) to obtain the polymer, preferably polymer A, polymer composition, preferably polymer composition A or polymer product, preferably polymer product A is / are described in more detail in paragraph

[2011] of Reference RF1.The term “industrial use polymer”, as used in the context of the product Q herein, comprises rhe-ology, polycarboxylate, alkoxylated polyalkylenamine, alkoxylated polyalkylenimine, poly- ether-based, dye inhibition and soil release cleaning polymers defined in more detail in paragraphs

[3035] to

[3044] of Reference RF1. The term “industrial use surfactant”, as used in the context of the product Q herein, comprises non-ionic, anionic and amphoteric industrial use surfactants defined in more detail in paragraphs

[3008] to

[3034] of Reference RF1. The term “industrial use descaling compound”, as used in the context of the product Q herein, comprises non-phosphate based builders (NPB) and phosphonates (CoP) described in more detail in paragraphs

[3001] to

[3005] of Reference RF1. The term “industrial use biocide”, as used herein, refers to a chemical compound that kills microorganisms or inhibits their growth or reproduction defined in more detail in paragraphs

[3006] to

[3007] of Reference RF1. The term “industrial use solvent”, as used in the context of the product Q herein, comprises alkyl amides, alkyl lactamides, alkyl esters, lactate esters, alkyl diester, cyclic alkyl diester, cyclic carbonates, aromatic aldehydes and aromatic esters defined in more detail in paragraphs

[3045] to

[3055] of Reference RF1 . The term “industrial use dispersant”, as used in the context of the product Q herein, comprises anionic and non-ionic industrial use dispersants defined in more detail in paragraphs

[3056] to

[3058] of Reference RF1 . The term “composition and / or formulation thereof” with reference to the industrial use polymers, industrial use surfactants, descaling compounds and / or industrial use biocides refers to industrial use compositions and / or institutional use products and / or fabric and home care products and / or personal care products defined in more detail in paragraph

[3059] of Reference RF1. The converting step(s) to obtain the industrial use polymer, industrial use surfactant, descaling compound and / or industrial use biocide are defined in more detail in paragraph

[3060] of Reference RF1. The converting steps to obtain the industrial use composition or formulation of the industrial use polymer, industrial use surfactant, descaling compound and / or industrial use biocide are defined in more detail in paragraph

[3061] of Reference RF1.The term “agrochemical composition”, as used in the context of the product Q herein, typically relates to a composition comprising an agrochemically active ingredient and at least one agrochemical formulation auxiliary. Examples of agrochemical compositions, active ingredients and auxiliaries are described in more detail in Reference RF1 , paragraph

[4001] ,The agrochemical composition may take the form of any customary formulation. The agrochemical compositions are prepared in a known manner, e.g. described by Mollet and Grube- mann, Formulation technology, Wiley VCH, Weinheim, 2001 ; or Knowles, New developments in crop protection product formulation, Agrow Reports DS243, T&F Informa, London, 2005. The converting step(s) to obtain the agrochemically active ingredients and auxiliaries may be conducted in analogy to the production step(s) of their analogues that are based on petrochemicals or other precursors that are not gained by recycling processes. In addition, conversion to compounds mentioned in sections “Polymer” and “Cosmetic surfactant, emollient, wax, cosmetic polymer, UV filter, further cosmetic ingredient or compositions or formulations thereof” may be performed as described in these sections as well as the respective paragraphs in Reference RF1.The term active pharmaceutical ingredients and / or intermediates thereof, as used in the context of the product Q herein, comprises substances that provide pharmacological activity or other direct effect in the diagnosis, cure, mitigation, treatment, or prevention of disease, or to affect the structure or any function of the body. Intermediates thereof are isolated products that are generated during a multi-step route of synthesis of an active pharmaceutical ingredient. The term pharmaceutical excipients, as used in the context of the product Q herein, comprises compounds or compound mixtures used in compositions for various pharmaceutical applications, which are not substantially pharmaceutically active on itself. Active pharmaceutical ingredients and / or intermediates thereof and pharmaceutical excipients are defined in more detail in paragraph

[5001] of Reference RF1.The converting step(s) to obtain the active pharmaceutical ingredients and / or intermediates thereof and pharmaceutical excipients may comprise one or more synthesis steps and can be performed by conventional synthesis and techniques well known to a person skilled in the art.The terms animal feed additives, human food additives, dietary supplements, as used in the con-text of the product Q herein, comprises Vitamins, Pro-Vitamins and active metabolites thereof including intermediates and precursors, especially Vitamin A, B, E, D, K and esters thereof, like acetate, propionate, palmitate esters or alcohols thereof like retinol or salts thereofand any combinations thereof; Tetraterpenes, especially isoprenoids like carotenoids and xanthophylls including their intermediates and precursors as well as mixtures and derivates thereof, especially beta carotene, Canthaxanthin, Citranaxanthin, Astaxanthin, Zeaxanthin, Lutein, Lycopene, Apo-carotenoids, and any combinations thereof; organic acids, especially formic acid, propionic acid and salts thereof, such as sodium, calcium or ammonium salts, and any combinations thereof, such as but not limited to mixtures of formic acid and sodium formiate, propionic acid and ammonium propionate, formic acid and propionic acid, formic acid and sodium formiate and propionic acid, propionic acid and sodium propionate and formic acid and sodium formiate; glycerides of carboxylic acids and short and medium chain fatty acids, conjugated linoleic acids, such as omega-6 fatty acid (C18:2) methyl ester and 1 ,2-propandiol and beverage stabilizers, such as polyvinylpyrrolidone-polymer or polyvinylimidazole / polyvinylpyrrolidone-co- polymer. Animal feed additives, human food additives and dietary supplements are defined in more detail in paragraph

[5002] of Reference RF1.The converting step(s) to obtain the animal feed additives, human food additives, dietary supplements may comprise one or more synthesis steps and can be performed by conventional synthesis and techniques well known to a person skilled in the art.The terms aroma chemical and aroma composition as used in the context of the product Q here-in, comprise a volatile organic substance with a molecular weight between 70-250 g / mol comprising a functional group with a carbon skeleton of C5-C16 carbon atoms comprising linear, branched, cyclic, for example with a ring size of C5-C18, bicyclic or tricyclic aliphatic chains and but not necessarily one or more unsaturated structural elements like double bonds, triple bonds, aromatics or heteroaromatics and preferably the one or more additional functional groups are selected from alcohol, ether, ester, ketone, aldehyde, acetal, carboxylic acid, nitrile, thiol, amine. In one aspect, the aroma chemical is a terpene-based aroma chemical, for example selected from monoterpenes and monoterpenoids, sesquiterpenes and sesquiterpenoids, diterpenes, triterpenes or tetraterpenes. Aroma chemicals can be combined with further aroma chemicals to give an aroma composition. Aroma chemicals and aroma compositions are defined in more de-tail in paragraph

[5003] of Reference RF1.The converting step(s) to obtain the aroma chemical and aroma composition may comprise one or more synthesis steps and can be performed by conventional synthesis and techniques well known to a person skilled in the art.The term “aqueous polymer dispersion”, as used in the context of the product Q herein, comprises aqueous composition(s) comprising dispersed polymer(s) and is defined in more detail inthe section

[6001] entitled “aqueous polymer dispersion” of Reference RF1. The dispersed polymers) may be selected from acrylic emulsion polymer(s), styrene acrylic emulsion polymer(s), styrene butadiene dispersion(s), aqueous dispersion(s) comprising composite particles, acrylate alkyd hybrid dispersion(s), polyurethane(s) (including UV-curable polyurethanes) and polyurethane - poly(meth)acrylate hybrid polymer(s). The term “emulsion polymer”, as used herein, comprises polymer(s) made by free-radical emulsion polymerization. Aqueous polyurethane dispersions) are defined in more detail in the section

[6002] entitled “Polyurethane dispersions” of Reference RF1. UV-curable polyurethane(s) is / are defined in more detail in the section

[6017] of Reference RF1. Polyurethane - poly(meth)acrylate hybrid polymer(s) is / are defined in more detail in the section

[6016] of Reference RF1.The term “polymeric dispersant”, as used in the context of the product Q herein, comprises preferably polymer(s) comprising polyether side chain, in particular polycarboxylate ether polymer(s) and polycondensation product(s) defined in more detail in paragraph

[6020] entitled “Polymeric dispersant” of Reference RF1.The converting (polymerization) step(s) to obtain the aqueous polymer dispersion(s) comprising emulsion polymer(s) is / are defined in more detail in the section

[6003] entitled “Emulsion polymerization” of Reference RF1.The converting (polymerization) step(s) to obtain the aqueous polyurethane dispersion(s) is / are defined in more detail in the section

[6014] entitled “Process for the preparation of aqueous poly-urethane dispersions” and section [6017)] entitled “Aqueous UV-curable polyurethane dispersions, their preparation and use and compositions containing them” of Reference RF1. Composition(s) and uses of aqueous polymer dispersion(s) and of polymeric dispersant(s) are defined in more detail in the following sections of Reference RF1 : section

[6004] entitled “Uses of aqueous polymer dispersions”, section

[6005] entitled “Binders for architectural and construction coatings” section

[6006] entitled “Binders for paper coating” section

[6007] entitled “Binders for fiber bonding” section

[6008] entitled “Adhesive polymers and adhesive compositions” section

[6015] entitled “Aqueous polyurethane dispersions suitable for use in coating compositions” section

[6016] entitled “Aqueous polyurethane - poly(meth)acrylate hybride polymer dispersions suitable for use in coating compositions” section

[6017] entitled “Aqueous UV-curable polyurethane dispersions, their preparation and use and compositions containing them”section

[6018] entitled “Inorganic binder compositions comprising polymeric dispersants and their use”

[6019] 100% curable coating compositionsUV-crosslinkable poly(meth)acrylate(s) and its / their uses are defined in more detail in section

[6009] entitled “UV-crosslinkable poly(meth)acrylates for use in UV-curable solvent-free hot melt adhesives and their use for making pressure-sensitive self-adhesive articles” of Reference RF1.Polyisocyanate(s), composition(s) comprising them and their uses are defined in more detail in section

[6010] entitled “Polyisocyanates” of Reference RF1.Hyperbranched polyester polyol(s) and its / their uses are defined in more detail in section

[6011] entitled “Organic solvent based hyperbranched polyester polyols suitable for use in coating com-positions” of Reference RF1. The converting step(s) to obtain the hyperbranched polyester polyols is / are defined in more detail in the section

[6012] entitled “Preparation of organic solvent based hyperbranched polyester polyols” of Reference RF1. Coating composition(s) comprising hyperbranched polyester polyol(s), polyisocyanate(s) and additive(s) and substrate(s) coated therewith are defined in more detail in section

[6013] entitled “Organic solvent based two component coating compositions comprising hyperbranched polyester polyols and polyisocyanates” of Reference RF1.Unsaturated polyester polyol(s), solvent-based coating composition(s) comprising said unsaturated polyester polyol(s) and substrate(s) for coating with said coating composition(s) are defined in more detail in section

[6018] entitled “Organic solvent based coating composition comprising unsaturated polyester polyols” of Reference RF1.100% curable coating composition(s) is / are defined in more detail in section

[6019] of Reference RF1.Polymeric dispersant(s) for inorganic binder compositions is / are defined in more detail in section

[6020] of Reference RF1. The inorganic binder composition(s) comprising the polymeric dispersants and their use are defined in more detail in section

[6021] of Reference RF1. The converting step(s) to obtain the polymeric dispersant(s) are defined in more detail in section

[6020] of Reference RF1. The term “inorganic binder composition” comprising the polymeric dispersants), as used herein, comprises preferably in particular hydraulically setting compositions and compositions comprising calcium sulfate and is defined in more detail in section

[6021] of Reference RF1 entitled “Inorganic binder compositions comprising the polymeric dispersant andtheir use”. Specific building material formulation(s) comprising polymeric dispersant(s) or building product(s) produced by a building material formulation comprising a polymeric dispersant are disclosed in more detail in section

[6021] of Reference RF1.The term “cosmetic surfactant”, as used in the context of the product Q herein, comprises nonionic, anionic, cationic and amphoteric surfactants and is defined in more detail in paragraph

[7002] of Reference RF1. The term “emollient”, as used in the context of the product Q herein, refers to a chemical compound used for protecting, moisturizing, and / or lubricating the skin and is defined in more detail in paragraph

[7003] of Reference RF1. The term “wax”, as used in the context of the product Q herein, comprises pearlizers and opacifiers and is defined in more detail in paragraph

[7004] of Reference RF1. The term “cosmetic polymer”, as used in the context of the product Q herein, comprises any polymer that can be used as an ingredient in a cosmetic formulation and is defined in more detail in paragraph

[7005] of Reference RF1. The term “UV filter”, as used in the context of the product Q herein, refers to a chemical compound that blocks or absorbs ultraviolet light and is defined in more detail in paragraph

[7006] of Reference RF1. The term “further cosmetic ingredient”, as used in the context of the product Q herein, comprises any ingredient suitable for making a cosmetic formulation. Several sources disclose cosmetically acceptable ingredients. E. g. the database Cosing on the internet pages of the European Com-mission discloses cosmetic ingredients and the International Cosmetic Ingredient Dictionary and Handbook, edited by the Personal Care Products Council (PCPC), discloses cosmetic ingredients. The term “composition and / or formulation thereof” with reference to the cosmetic surfactant, emollient, wax, cosmetic polymer, UV filter and / or further cosmetic ingredient refers to personal care and / or cosmetic compositions or formulations defined in more detail in paragraph

[7007] of Reference RF1. The converting step(s) to obtain the cosmetic surfactant, emollient, wax, cosmetic polymer, UV filter or further cosmetic ingredient is / are defined in more detail in paragraph

[7008] of Reference RF1.The terms “polymer B”, “polymer composition B”, “coating composition”, “other functional composition”, “foil”, “molded body”, “coating” and “coated substrate” are well known to the person skilled in the art and are defined in more detail from paragraph

[8000] to

[8005] of Reference RF1.The present invention is further illustrated by the following set of embodiments and combinations of embodiments resulting from the dependencies and back-references as indicated. In particular, it is noted that in each instance where a range of embodiments is mentioned, for example in the context of a term such as "The process of any one of embodiments 1 to 3", every embodiment in this range is meant to be explicitly disclosed for the skilled person, i.e. the wordingof this term is to be understood by the skilled person as being synonymous to "The process of any one of embodiments 1 , 2 and 3". Further, it is explicitly noted that the following set of embodiments represents a suitably structured part of the general description directed to preferred aspects of the present invention, and, thus, suitably supports, but does not represent the claims of the present invention but disclose exemplary processes, methods, sorting units, models, computer programs, computer-readable storage mediums and uses that form part of the description.1 . A process for sorting a plastic-containing material, the process comprising(i) providing a particulate plastic-containing material Ms;(ii) subjecting the particulate material Ms according to (i) to sorting in a sorting unit Us, said sorting unit Us comprising(a) a classification section Sc for classifying particles of the particulate material Ms with respect to their polyamide 6 content and polyamide 66 content, said classification section Sc comprising(a.1) an optical system configured for measuring at least two wavelengths and determining spectroscopic data of the particles of the particulate material Ms therefrom;(a.2) a processing unit UP configured for discriminating polyamide 6 comprised in a particle of Ms from polyamide 66 comprised in said particle of Ms by performing a computer-implemented classification method using the spectroscopic data according to (a.1);(b) a sorting section Ss comprising a sorting device Ds configured for sorting particles of Ms classified in Sc according to their respective polyamide 6 and polyamide 66 content; wherein subjecting the particulate material Ms to sorting comprises(ii.1 ) feeding the particulate material Ms provided according to (i) to the classification section Sc according to (a);(11.2) determining spectroscopic data of x % of the particles of Ms with the optical system according to (a.1), comprising measuring at least two wavelengths of a given particle in a wavelength range Rw of from 0.9 to 12 pm, wherein x > 50;(11.3) classifying the particles spectroscopically examined according to (ii.2) in the processing UP according to (a.2), comprising performing the computer-implemented classification method using the spectroscopic data determined according to (ii.2), obtaining particles classified according to their polyamide 6 and polyamide 66 content;(ii.4) sorting the particles classified according to (ii.3) according to their respective polyamide 6 and polyamide 66 content in Ss via the sorting device Ds, obtaining a sorted particulate plastic-containing material Ms.2. The process of embodiment 1 , wherein the wavelength range Rw according to (ii.2) is of from 3 to 12 pm, preferably of from 4 to 12 pm, preferably of from 5 to 12 pm, more preferably of from 6 to 12 pm, more preferably of from 7 to 12 pm, more preferably of from 8 to 12 pm.3. The process of embodiment 1 or 2, wherein x according to (ii.2) is in the range of from 50 to 100, preferably in the range of from 70 to 100, more preferably in the range of from 90 to 100, more preferably in the range of from 95 to 100.4. The process of any one of embodiments 1 to 3, wherein the particulate plastic-containing material Ms comprises, preferably consists of a particulate plastic-containing waste material, more preferably of a particulate engineering plastic-containing waste material.5. The process of any one of embodiments 1 to 4, wherein the particulate plastic-containing material Ms comprises one or more of polyamide 6 and polyamide 66 and preferably further comprises carbon black.6. The process of any one of embodiments 1 to 5, wherein from 50 to 100 weight-%, preferably from 70 to 100 weight-%, more preferably from 80 to 100 weight-%, more preferably from 90 to 100 weight-%, more preferably from 95 to 100 weight-% of the particulate plastic-containing material Ms consists of one or more of polyamide 6 and polyamide 66 and preferably additionally carbon black.7. The process of any one of embodiments 1 to 6, wherein less than 100 weight-% of the particulate plastic-containing material Ms consist of one or more of polyamide 6 and polyamide 66 and preferably additionally carbon black, the plastic-containing material Ms further comprising one or more components selected from the group consisting of at least one polymer compound other than polyamide 6 and polyamide 66, wood, glass, and at least one metal, wherein the at least one polymer compound preferably comprises one or more of polypropylene, polycarbonate, polydimethylsiloxane, polyethylene, polyurethane, acrylonitrile-butadiene-styrene, acrylonitrile-butadiene-styrene, acrylonitrile styrene acrylate, polyethylene terephthalate, polyester, polystyrene, polyvinylchloride, a copolymer oftwo or more thereof including statistical copolymers, gradient copolymers, alternating copolymers, block copolymers, and graft copolymers, polymer blends of two or more thereof, a rubber material comprising one or more of a natural rubber material and a synthetic rubber material. The process of any one of embodiments 1 to 7, wherein providing the particulate material Ms according to (i) comprises(1.1) shredding a vehicle comprising polymeric vehicle parts, obtaining an automotive shredder residue (ASR);(1.2) preferably subjecting the automotive shredder residue according to (i.1) to one or more post-treatment steps (i.2.1) to (i.2.3):(i.2.1 ) separating metal fragments, including, but not restricted to one or more of ferrous metal fragments and non-ferrous metal fragments from the automotive shredder residue;(i.2.2.) separating the automotive shredder residue into a shredder light fraction and a shredder heavy fraction;(i.2.3) subjecting the automotive shredder residue, preferably one or more of the the shredder light fraction and the shredder heavy fraction according to(i.2.2) to an aqueous treatment, said aqueous treatment preferably comprising one or more of washing and aqueous density separation; wherein the particulate material Ms is obtained from (i.1), preferably from (i.2). The process of any one of embodiments 1 to 7, wherein the particulate material Ms provided according to (i) is obtainable or obtained from a method comprising(1.1) shredding a vehicle comprising polymeric vehicle parts, obtaining an automotive shredder residue (ASR);(1.2) preferably subjecting the automotive shredder residue according to (i.1) to one or more post-treatment steps (i.2.1) to (i.2.3):(i.2.1 ) separating metal fragments, including, but not restricted to one or more of ferrous metal fragments and non-ferrous metal fragments from the automotive shredder residue;(i.2.2.) separating the automotive shredder residue into a shredder light fraction and a shredder heavy fraction;(i.2.3) subjecting the automotive shredder residue, preferably one or more of the the shredder light fraction and the shredder heavy fraction according to(i.2.2) to an aqueous treatment, said aqueous treatment preferably comprising one or more of washing and aqueous density separation;wherein the particulate material Ms is obtained from (i.1), preferably from (i.2). The process of any one of embodiments 1 to 9, wherein performing the computer-implemented classification according to (ii.3) comprises(a) providing a computer-implemented trained classification model;(P) retrieving the trained model provided according to (a);(y) retrieving spectroscopic data of a particle of Ms determined according to (ii.2);(5) applying the trained model to the spectroscopic data according to (y), thereby discriminating polyamide 6 comprised from polyamide 66 and obtaining particles classified according to their polyamide 6 and polyamide 66 content. The process of embodiment 10, wherein providing the computer-implemented trained classification model according to (a) comprises(a.1) providing a computer-implemented trainable model;(a.2) training the trainable model provided according to (a.1) with a computer-implemented training method, the training method comprising(a.2.1) retrieving a labelled spectroscopic training data set comprising spectroscopic data of polymer compounds determined by an optical system configured for measuring at least two wavelengths in the wavelength range Rw of polymer compounds having defined chemical composition;(a.2.2) training the trainable model on the labelled spectroscopic training data set, preferably by supervised training. The process of embodiment 11 , wherein the polymer compounds having defined chemical composition include a polymer compound consisting of polyamide 6, a polymer compound consisting of polyamide 66, optionally include polymer compounds consisting of polyamide 6 and polyamide 66 optionally at different respective amounts, and further optionally include polymer compounds consisting of one or more of polyamide 6 and polyamide 66 and a further polymer component preferably selected from the group consisting of polypropylene, polycarbonate, polydimethylsiloxane, polyethylene, polyurethane, acrylonitrile- butadiene-styrene, acrylonitrile-butadiene-styrene, acrylonitrile styrene acrylate, polyethylene terephthalate, polyester, polystyrene, polyvinylchloride, a copolymer of two or more thereof including statistical copolymers, gradient copolymers, alternating copolymers, block copolymers, and graft copolymers, polymer blends of two or more thereof, a rubber material comprising one or more of a natural rubber material and a synthetic rubber material, and mixtures of two or more thereof.13. The process of embodiment 11 or 12, wherein the trainable model according to (a.1) comprises at least one model selected from the group consisting of a decision tree model, preferably at least one of an XGBoost model and a Random Forest model; a Support Vector Machine (SVM) model, preferably a linear kernel SVM; a nearest neighbors model, preferably a KNeighbors Classifier; a Bayes model, preferably a GaussianNB() model; a regression model, preferably at least one of a LogisticRegression, a linear and a nonlinear regression model including a regression model comprising transformed features which includes one or more of a log-transformed and polynomial model including a Partial Least Squares Discriminant Analysis (PLS-DA) model; an Artificial Neural Network (ANN), preferably a non-linear Artificial Neural Network (ANN), more preferably a deep learning architecture including one or more of Convolutional NN, Recurrent NN, a Long Short Term Memory NN; a kernel-based method; a tree regression model; a distributed gradientboosting framework model, preferably a light gradient-boosting machine model (LightGBM classifier); a soft independent modelling by class analogy (SIMCA); a Multi Curve Resolution (MCR) model; a t-distributed stochastic neighbor embedding (t-SNE) model; a Uniform Manifold Approximation and Projection for Dimension Reduction (UMAP) model; a Classical Least Squares (CLS) model; a Simultaneous Component Analysis (ASCA- ANO A) model; and a Multi-Level Simultaneous Component Analysis (MLSCA) model.14. The process of any one of embodiments 11 to 13, wherein the spectroscopic data according to (a.2.1) are determined with an optical system configured for measuring two or more wavelengths, preferably three or more wavelengths, more preferred five or more wavelengths, in particular a hyperspectral camera CH having a spectral resolution of 150 to 200 spectral bands, preferably in the wavelength range Rw as defined in embodiments 1 and 2.15. The process of any one of embodiments 11 to 14, wherein the training method according to (a.2) further comprises pre-processing the spectroscopic data determined by CH to obtain the spectroscopic training data set, the pre-processing preferably comprising applying a Standard Normal Variate (SNV) to the spectroscopic data obtained.16. The process of embodiment 15, wherein the pre-processing further comprises taking at least one derivative of the spectroscopic data determined by CH, preferably of the SNV pre-processed spectroscopic data.17. The process of embodiment 15 or 16, wherein the pre-processing further comprises normalizing the spectroscopic data determined by CH, preferably normalizing the at least onederivative of the spectroscopic data, more preferably normalizing the at least one derivative of the SNV pre-processed spectroscopic data.18. The process of any one of embodiments 11 to 17, wherein the training method according to (a.2) further comprises at least one cross validation in the spectroscopic training data set to obtain at least one pair of training data set and test data set.19. The process of embodiment 18, wherein the training method further comprises at least one validation step comprising testing the accuracy of the training by using the test data set.20. The process of any one of embodiments 10 to 19, wherein applying the trained model according to (5) comprises attributing a classification probability to a given particle of Ms, wherein the particle is classified as a polyamide 6 particle if the polyamide 6 classification probability is at least 0.6, preferably at least 0.7, more preferably at least 0.8, more preferably at least 0.9, and wherein the particle is classified as a polyamide 66 particle if the polyamide 66 classification probability is at least 0.6, preferably at least 0.7, more preferably at least 0.8, more preferably at least 0.9.21 . The process of embodiment 20, wherein the trained model is a trained Partial Least Squares Discriminant Analysis (PLS-DA) model, wherein the classification probability of the given particle of Ms is determined by using classification results obtained by applying the PLS-DA model to the spectroscopic data of the particle of Ms.22. The process of any one of embodiments 1 to 21 , wherein the sorting device Ds comprised in the sorting section Ss comprises three outlet devices DOPA6, DOPA66 and DON, and wherein sorting the classified particles according to (ii.4) comprises controlling the sorting device Ds so that a particle classified as a polyamide 6 particle is passed to DOPA6, a particle classified as a polyamide 66 particle is passed to DOPA66, and a particle classified neither as a polyamide 6 particle nor as a polyamide 66 particle is passed to DON; wherein the sorting device Ds preferably comprises a first sorting subdevice Dssi, preferably a first pneumatic sorting subdevice or a first mechanical sorting subdevice, configured for actively removing a particle classified as a polyamide 6 particle to DOPA6 from the particulate material Ms or for actively removing a particle classified as a polyamide 66 particle to DOPA66 from the particulate material Ms; anda second sorting subdevice Dss2, preferably a second pneumatic sorting subdevice or a second mechanical sorting subdevice, configured for actively removing a particle classified as a polyamide 66 particle to DOPA66 or for actively removing a particle classified as a polyamide 6 particle to DOPA6 from the remainder material. The process of embodiment 22, wherein the particulate material Ms provided according to (i) comprises particles classified as polyamide 6 particles according to (ii.3), particles classified as polyamide 66 particles according to (ii.3), and optionally particles classified neither as polyamide 6 particles nor as polyamide 66 particles, the process comprising(11.4.1) actively removing particles classified as polyamide 6 particles to DOPA6 from the particulate material Ms via Dssi, obtaining sorted-out particles classified as polyamide 6 particles and further obtaining a first remainder material MSRI comprising particles classified as polyamide 66 particles and optionally particles neither classified as polyamide 66 particles nor classified as polyamide 6 particles;(11.4.2) actively removing particles classified as polyamide 66 particles to DOPA66 from the remained material according to (ii.4.1) via Dss2, obtaining sorted-out particles classified as polyamide 66 particles and optionally further obtaining a second remainder material MSR2 comprising particles classified neither as polyamide 66 particles nor as polyamide 6 particles, said second remainder stream being passed to DON; or(ii.4.1’) actively removing particles classified as polyamide 66 particles to DOPA66 from the particulate material Ms via Dssi, obtaining sorted-out particles classified as polyamide 66 particles and further obtaining a first remainder material MSRI comprising particles classified as polyamide 6 particles and optionally particles neither classified as polyamide 66 particles nor classified as polyamide 6 particles;(ii .4.2’) actively removing particles classified as polyamide 6 particles to DOPA6 from the remained material according to (ii.4.1) via Dss2, obtaining sorted-out particles classified as polyamide 6 particles and optionally further obtaining a second remainder material MSR2 comprising particles classified neither as polyamide 6 particles nor as polyamide 66 particles, said second remainder stream being passed to DON. The process of embodiment 22 or 23, wherein controlling Ds, and preferably Dssi and DSS2, is computer-controlled. The process of any one of embodiments 1 to 24, further comprising(iii) passing sorted plastic-containing material to recycling, said recycling comprising one or more of (iii.1) and (iii.2):(111.1) passing sorted plastic-containing material classified as polyamide 6 plastic material, preferably sorted-out particles classified as polyamide 6 particles obtained according to (ii.4.1) or (ii.4.2’) according to embodiment 22 to one or more of depolymerization, including one or more of neutral, acidic and alkaline depolymerization, and enzymatic depolymerization; solvolysis, preferably to solvolysis with one or more alcohols, more preferably with methanol; solvent-based recycling;(111.2) passing sorted plastic-containing material classified as polyamide 66 plastic material, preferably sorted-out particles classified as polyamide 66 particles obtained according to (ii.4.2) or (ii.4.1’) according to embodiment 22 to one or more of depolymerization, including one or more of neutral, acidic and alkaline depolymerization, and enzymatic depolymerization; solvolysis, preferably to solvolysis with one or more alcohols, more preferably with methanol; solvent-based recycling. A process for recycling plastic-containing waste material, comprising sorting plastic-containing waste material according to a process according to any one of embodiments 1 to 24, the recycling process further comprising(iii) passing sorted plastic-containing material to recycling, said recycling comprising one or more of (iii.1) and (iii.2):(111.1) passing sorted plastic-containing material classified as polyamide 6 plastic material, preferably sorted-out particles classified as polyamide 6 particles obtained according to (ii.4.1) or (ii.4.2’) according to embodiment 22 to one or more of depolymerization, including one or more of neutral, acidic and alkaline depolymerization, and enzymatic depolymerization; solvolysis, preferably to solvolysis with one or more alcohols, more preferably with methanol; solvent-based recycling;(111.2) passing sorted plastic-containing material classified as polyamide 66 plastic material, preferably sorted-out particles classified as polyamide 66 particlesobtained according to (ii .4.2) or (ii.4.1 ’) according to embodiment 22 to one or more of depolymerization, including one or more of neutral, acidic and alkaline depolymerization, and enzymatic depolymerization; solvolysis, preferably to solvolysis with one or more alcohols, more preferably with methanol; solvent-based recycling. A process for recycling plastic-containing waste material, comprising(iii) passing sorted plastic-containing material to recycling, said sorted plastic-containing material being obtainable or obtained by a a process according to any one of embodiments 1 to 24; said recycling comprising one or more of (iii.1) and (iii.2):(111.1) passing sorted plastic material classified as polyamide 6 plastic material, preferably sorted-out particles classified as polyamide 6 particles obtained according to (ii.4.1 ) or (ii.4.2’) according to embodiment 22 to one or more of depolymerization, including one or more of neutral, acidic and alkaline depolymerization, and enzymatic depolymerization; solvolysis, preferably to solvolysis with one or more alcohols, more preferably with methanol; solvent-based recycling;(111.2) passing sorted plastic-containing material classified as polyamide 66 plastic material, preferably sorted-out particles classified as polyamide 66 particles obtained according to (ii.4.2) or (ii.4.1 ’) according to embodiment 22 to one or more of depolymerization, including one or more of neutral, acidic and alkaline depolymerization, and enzymatic depolymerization; solvolysis, preferably to solvolysis with one or more alcohols, more preferably with methanol; solvent-based recycling. A sorting unit Us for sorting a plastic-containing material Ms, preferably for carrying out a sorting process according to any one of embodiments 1 to 27, the sorting unit Us comprising(a) a classification section Sc for classifying particles of the particulate material Ms with respect to their polyamide 6 content and polyamide 66 content, said classification section Sc comprising(a.1) an optical system configured for measuring at least two wavelengths and determining spectroscopic data of the particles of the sorting material therefrom;(a.2) a processing unit UP configured for discriminating polyamide 6 comprised a particle of Ms from polyamide 66 comprised in said particle of Ms by performing a computer-implemented classification method using the spectroscopic data according to (a.1);(b) a sorting section Ss comprising a sorting device Ds configured for sorting particles of Ms classified in Sc according to their respective polyamide 6 and polyamide 66 content, the sorting section Ss preferably further comprising three outlet devices DOPA6, DQPA66 and DON.29. The sorting unit of embodiment 28, wherein the sorting device Ds comprises a first sorting subdevice Dssi, preferably a first pneumatic sorting subdevice or a first mechanical sorting subdevice, configured for actively removing a particle classified as a polyamide 6 particle to DOPA6 from the particulate material Ms or for actively removing a particle classified as a polyamide 66 particle to DOPA66 from the particulate material Ms; and a second sorting subdevice Dss2, preferably a second pneumatic sorting subdevice or a second mechanical sorting subdevice, configured for actively removing a particle classified as a polyamide 66 particle to DOPA66 or for actively removing a particle classified as a polyamide 6 particle to DOPA6 from the remainder material.30. The sorting unit of embodiment 28 or 29, wherein controlling Ds is computer-controlled.31 . A computer-implemented classification method for discriminating polyamide 6 from polyamide 66, preferably for discriminating polyamide 6 comprised in a particle of a particulate plastic-containing material Ms from polyamide 66 comprised in said particle of Ms, said computer-implemented classification method comprising(a) providing a computer-implemented trained classification model;(P) retrieving the trained model provided according to (a);(y) retrieving spectroscopic data of polymer compounds comprising one or more of polyamide 6 and polyamide 66, said spectroscopic data being determined by an optical system configured for measuring at least two wavelengths in a wavelength range Rw of from 0.9 to 12 pm;(5) applying the trained model to the spectroscopic data according to (y), thereby discriminating polyamide 6 from polyamide 66;wherein Rw according to (y) is preferably of from 3 to 12 pm, more preferably of from 4 to 12 pm, more preferably of from 5 to 12 pm, more preferably of from 6 to 12 pm, more preferably of from 7 to 12 pm, more preferably of from 8 to 12 pm.32. The classification method of embodiment 31 , wherein providing the trained model trained by using at least one computer-implemented training method according to (a) comprises (a.1) providing a computer-implemented trainable model;(a.2) training the trainable model provided according to (a.1) with a computer-implemented training method, the training method comprising(a.2.1) retrieving a labelled spectroscopic training data set, the spectroscopic training data set comprising spectroscopic data of polymer compounds determined by an optical system configured for measuring at least two wavelengths in the wavelength range Rw of polymer compounds having defined chemical composition;(a.2.2) training the trainable model on the labelled spectroscopic training data set, preferably by supervised training.33. The classification method of embodiment 31 or 32, being carried out in step (ii.3) of a sorting process according to any one of embodiments 1 to 25 or in a process according to embodiment 26 or 27.34. A method for providing a trained model for discriminating polyamide 6 from polyamide 66, preferably for discriminating polyamide 6 comprised a particle of a particulate plastic-containing material Ms from polyamide 66 comprised in said particle of Ms, said method comprising(a.1) providing a computer-implemented trainable model;(a.2) training the trainable model provided according to (a.1) with a computer-implemented training method, the training method comprising(a.2.1) retrieving a labelled spectroscopic training data set, the spectroscopic training data set comprising spectroscopic data of polymer compounds determined by an optical system configured for measuring at least two wavelengths in a wavelength range Rw of from 0.9 to 12 pm of polymer compounds having defined chemical composition;(a.2.2) training the trainable model on the labelled spectroscopic training data set, preferably by supervised training;wherein Rw according to (a.2.1) is preferably of from 3 to 12 pm, more preferably of from 4 to 12 pm, more preferably of from 5 to 12 pm, more preferably of from 6 to 12 pm, more preferably of from 7 to 12 pm, more preferably of from 8 to 12 pm.35. The method of embodiment 34, being carried out in step (a) of a computer-implemented classification method according to any one embodiments 31 to 33.36. A trained model for discriminating polyamide 6 from polyamide 66, provided by a method according to embodiment 34 or 35.37. A computer program, comprising instructions which, when the program is executed by at least one processor of a computer or a computer network, cause the processor to perform the training method according to embodiment 34 or 35.38. A computer program, comprising instructions which, when the program is executed by at least one processor of a computer or a computer network, cause the processor to perform the classification method according to any one of embodiments 31 to 33.39. A computer-readable storage medium, preferably a non-transient computer-readable storage medium, comprising instructions which, when the instructions are executed by at least one processor of a computer or a computer network, cause the processor to perform the training method according to embodiment 34 or 35.40. A computer-readable storage medium, preferably a non-transient computer-readable storage medium, comprising instructions which, when the instructions are executed by at least one processor of a computer or a computer network, cause the processor to perform the classification method according to any one embodiments 31 to 33.41. Sorted plastic-containing material, preferably sorted plastic-containing waste material, obtainable or obtained by a process according to any one of embodiments 1 to 25.42. A method for recycling plastic-containing material, comprising providing sorted plasticcontaining material, preferably sorted plastic-containing waste material, said sorted material being obtainable or obtained by a process according to any one of embodiments 1 to 25 and / or by using the sorting unit according to any one of embodiments 28 to 30, and recycling said sorted plastic-containing material to obtain a polymer; wherein the method forrecycling preferably comprises, more preferably consists of one or more of mechanical recycling, chemical recycling and solvent-based recycling. Use of sorted plastic-containing material, preferably sorted plastic-containing waste material, in a recycling process, wherein the sorted material is obtainable or obtained by a process according to any one of embodiments 1 to 25 and / or by using the sorting unit according to any one of embodiments 28 to 31 . A process, preferably according to any one of embodiments 1 to 25, comprising the step of converting a chemical material obtainable or obtained by the process according to any one of embodiments 1 to 25 to obtain a product Q. A process, comprising the step of using the sorting unit according to any one of embodiments 28 to 31 to obtain a chemical material; and preferably converting the chemical material to obtain a product Q. The process of embodiment 44 or 45, wherein the product Q is selected from: building block or monomer; or polymer, preferably polymer A, polymer composition, preferably polymer composition A, or polymer product, preferably polymer product A; or industrial use polymer, industrial use surfactant, descaling compound, industrial use biocide, industrial use solvent, industrial use dispersant, composition thereof or formulation thereof; or agrochemical composition, agrochemical formulation auxiliary or agrochemically active ingredient; or active pharmaceutical ingredient or intermediate thereof, pharmaceutical excipient, animal feed additive, human food additive, dietary supplements, aroma chemical or aroma composition; or aqueous polymer dispersion, preferably polyurethane or polyurethane - poly(meth) acrylate hybrid polymer dispersion, emulsion, binder for paper and fiber coatings, UV-curable acrylic polymer for hot melts and coatings polyisocyanates, hyperbranched polyester polyol, polymeric dispersant for inorganic binder compositions, unsaturated polyester polyol or 100% curable composition; or cosmetic surfactant, emollient, wax, cosmetic polymer, UV filter, further cosmetic ingredient or composition or formulation thereof; or polymer B, polymer composition B, coating composition, other functional composition, foil, molded body, coating or coated substrate.47. The process of any one of embodiments 44 to 46, wherein the content of the chemical material in the product Q is 1 weight-% or more, preferably 2 weight-% or more, more preferably 5 weight-% or more, more preferably 15 weight-% or more, more preferably 30 weight-% or more, more preferably 40 weight-% or more, more preferably 60 weight-% or more, more preferably 80 weight-% or more, more preferably 90 weight-% or more, more preferably 95 weight-% or more; and / or wherein the content of the chemical material in the product Q is 100 weight-% or less, preferably 95 weight-% or less, more preferably 90 weight-% or less, more preferably 50 weight-% or less, more preferably 25 weight-% or less, more preferably 10 weight-% or less; and preferably wherein the content is determined based on identity preservation and / or segregation and / or mass balance and / or book and claim chain of custody models, preferably based on mass balance, preferably the International Sustainability and Carbon Certification (ISCC) standard.It is explicitly noted that the above set of embodiments represents a suitably structured part of the general description directed to preferred aspects of the present invention, and, thus, suitably supports, but is not the set of claims of the present invention.The present invention is further illustrated by the following example.ExamplesExample: Training methodIn a first example, spectroscopic data were acquired using a Specim® FX120 hyperspectral camera as optical system configured for measuring two or more wavelengths, preferably three or more wavelengths, more preferred five or more wavelengths. Spectroscopic data from different material types of PA6 and PA66 were collected for two different types of samples. The two samples were analyzed with the hyperspectral camera as the optical system. A first sample was a pure PA6 sample and a second sample was a pure PA66 sample. The samples were particulate material samples. The spectroscopic data were obtained from measuring two or more wavelengths, preferably three or more wavelengths, more preferred five or more wavelengths with the optical system, in particular hyperspectral images of each sample comprised 3255 pixels for the PA6 sample and 3432 pixels for the PA66 sample, from which spectroscopic data in the wavelength range between 7924.25 nm to 10841.4 nm were obtained. The trainable model was a Partial Least Squares Discriminant Analysis (PLS-DA) model. The raw spectroscopic data obtained with theoptical system, in particular from the hyperspectral images, were pre-processed using SNV and a subsequent 1st-order derivative (order: 2, window: 3 pt, tails: weighted). The spectroscopic data were subsequently normalized with a 1-Norm and a normed area of 1. The spectral variable included was on the range between 7924.25 nm to 10841.4 nm. A number of four latent variables was used for the PLS-DA model representing 82,5% of the total variance explained. The training process included a cross validation, wherein the cross validation method was Venetian blinds with 10 splits and blind thickness of 1.The training results are shown in the following Tables 1 to 6. In Tables 1 to 6, the following ab- breviations are used:Cal CalibrationVai ValidationCV Cross ValidationRMSEC Root Mean Squared Error for CalibrationRMSECV Root Mean Squared Error for Cross ValidationTPR True Positive RateFPR False Positive RateTNR True Negative RateFNR False Negative RateF1 F1 ScoreTable 1Model Statistics for Example 1Table 2Percent Variance Capture by the trainable model on each componentTable 3 Confusion matrix for trainingTable 4Confusion matrix for cross validationTable 5Confusion table for trainingTable 6Confusion table for cross validationShort description of the FiguresFurther optional features and embodiments will be disclosed in more detail in the subsequent description of embodiments, preferably in conjunction with the dependent claims. Therein, the respective optional features may be realized in an isolated fashion as well as in any arbitrary feasible combination, as the skilled person will realize. The scope of the invention is not restricted by the preferred embodiments. The embodiments are schematically depicted in the Figures. Therein, identical reference numbers in these Figures refer to identical or functionally comparable elements.In the Figures:Figure 1 shows a diagram with spectroscopic data of training samples;Figure 2 shows a diagram with a comparison between ATR-FTIR spectroscopic data and spectroscopic data obtained from the hyperspectral camera as optical system;Figure 3 shows a diagram of the variance importance in projection of a trained model;Figure 4 shows a diagram of the selectivity ratio of a trained model;Figures 5A to 50 show RGB images and heat maps of classification probabilities for PA6 test samples;Figures 6A to 60 show RGB images and heat maps of classification probabilities for PA66 test samples;Figure 7 shows diagrams with spectroscopic data (Figure 7A) and selectivity ratios(Figure 7B) of PA6 test samples;Figure 8 shows diagrams with spectroscopic data (Figure 8A) and selectivity ratios(Figure 8B) of PA66 test samples; andFigure 9 shows diagrams with spectroscopic data (Figure 9A) and selectivity ratios(Figure 9B) of test samples comprising other polymers than PA6 and PA66.Detailed description of the FiguresFigures 1 to 60 refer to Example 1 . Specifically, Figures 1 to 4 show training results of the training process of Example 1 , whereas Figures 5A to 60 show prediction results using the trained PLS-DA model being trained according to Example 1.Figure 1 shows a diagram with spectroscopic data of the training samples in Example 1 . Specifically, the overlapped mean spectra 110 of the training samples in arbitrary units are shown as a function of wavenumber 112 in cm-1. The black dashed line represents the spectroscopic data of the PA6 sample obtained via an optical system configured for measuring two or more wavelengths, preferably three or more wavelengths, more preferred five or more wavelengths, in particular the hyperspectral camera (denoted by reference number 114). The gray solid line represents the spectroscopic data of the PA66 sample obtained via the hyperspectral camera (denoted by reference number 116). The diagram of Figure 1 shows the input variables used to train the PLS-DA model. As can be seen in Figure 1 , it is possible to observe spectral difference between the PA6 sample and the PA66 sample in the spectroscopic data obtained via the hyperspectral camera.Figure 2 shows a diagram with a comparison between attenuated total reflection-Fourier transform infrared spectroscopy (ATR-FTIR) spectroscopic data and spectroscopic data obtained from the hyperspectral camera. Specifically, the overlapped mean spectra 110 are shown as a function of wavenumber 112. The black bold line represents the analytical ATR-FTIR spectroscopic data for the PA6 sample (denoted by reference number 118) and the gray bold line represents the analytical ATR-FTIR spectroscopic data for the PA66 sample (denoted by reference number 120). As in Figure 1 , the black dashed line represents the spectroscopic data of the PA6 sample obtained via the hyperspectral camera (denoted by reference number 114) and the gray solid line represents the spectroscopic data of the PA66 sample obtained via the hyperspectral camera (denoted by reference number 116). As can be seen in Figure 2, there is a high correlation between the spectroscopic data of the ATR-FTIR and the spectroscopic data of the hyperspectral camera for both types of samples. However, using an optical system which isconfigured for measuring two or more wavelengths, preferably three or more wavelengths, more preferred five or more wavelengths, may also be feasible.Figure 3 shows a diagram of the variance importance in projection (VIP) 122 of the trained PLS- DA model according to Example 1. In Figure 3, the black dashed line represents the VIP for the PA6 sample (denoted by reference number 124) and the gray solid line represents the VIP for the PA66 sample (denoted by reference number 126).Figure 4 shows a diagram of the selectivity ratio 128 of the trained model PLS-DA model according to Example 1. In Figure 4, the black dashed line represents the selectivity ratio for the PA6 sample (denoted by reference number 130) and the gray solid line represents the selectivity ratio for the PA66 sample (denoted by reference number 132).Figures 5A to 50 show RGB images and heat maps of classification probabilities for PA6 test samples using the trained PLS-DA model according to Example 1. Specifically, Figures 5A, 5D, 5G, 5J and 5M show RGB images of the PA6 test samples. The remaining Figures show heat maps of the classification probabilities. For the test sample of Figure 5A, Figure 5B shows the classification probability of the test sample to be a polyamide 6 sample and Figure 5C shows the classification probability of the test sample to be a polyamide 66 sample. For the test sample of Figure 5D, Figure 5E shows the classification probability of the test sample to be a polyamide 6 sample and Figure 5F shows the classification probability of the test sample to be a polyamide 66 sample. For the test sample of Figure 5G, Figure 5H shows the classification probability of the test sample to be a polyamide 6 sample and Figure 5I shows the classification probability of the test sample to be a polyamide 66 sample. The test sample of Figure 5G is a test plate and is to be scratched to reduce the reflectivity of the surface of the test plate. For the test sample of Figure 5J, Figure 5K shows the classification probability of the test sample to be a polyamide 6 sample and Figure 5L shows the classification probability of the test sample to be a polyamide 66 sample. For the test sample of Figure 5M, Figure 5N shows the classification probability of the test sample to be a polyamide 6 sample and Figure 50 shows the classification probability of the test sample to be a polyamide 66 sample. As can be seen in Figures 5A to 50, the test samples can be determined accurately, i.e. with a high classification probability of above 0.7, to be PA6 samples by using the trained PLS-DA model of Example 1.Figures 6A to 60 show RGB images and heat maps of classification probability for PA66 test samples using the trained PLS-DA model according to Example 1. Specifically, Figures 6A, 6D, 6G, 6J and 6M show RGB images of the PA6 test samples. The remaining Figures show heat maps of the classification probabilities. For the test sample of Figure 6A, Figure 6B shows theclassification probability of the test sample to be a polyamide 6 sample and Figure 6C shows the classification probability of the test sample to be a polyamide 66 sample. For the test sample of Figure 6D, Figure 6E shows the classification probability of the test sample to be a polyamide 6 sample and Figure 6F shows the classification probability of the test sample to be a polyamide 66 sample. For the test sample of Figure 6G, Figure 6H shows the classification probability of the test sample to be a polyamide 6 sample and Figure 6I shows the classification probability of the test sample to be a polyamide 66 sample. The test sample of Figure 6G is a test plate and is to be scratched to reduce the reflectivity of the surface of the test plate. For the test sample of Figure 6J, Figure 6K shows the classification probability of the test sample to be a polyamide 6 sample and Figure 6L shows the classification probability of the test sample to be a polyamide 66 sample. For the test sample of Figure 6M, Figure 6N shows the classification probability of the test sample to be a polyamide 6 sample and Figure 60 shows the classification probability of the test sample to be a polyamide 66 sample. As can be seen in Figures 6A to 60, the test samples can be determined accurately, i.e. with a high classification probability of above 0.7, to be PA66 samples using the trained PLS-DA model of Example 1.Figure 7A shows a diagram illustrating spectroscopic data as a course of a signal over wavenumber (cm-1) for mean spectra of polyamide 6 (Pa6) samples. These and all other spectroscopic data as presented herein have been recorded by using a hyperspectral camera CH having a spectral resolution of 150 to 200 spectral bands, preferably in a wavelength range Rw of 7800 cm-110500 cm-1as indicated in the respective Figures. As an alternative, any other optical system configured for measuring two or more wavelengths, preferably three or more wavelengths, more preferred five or more wavelengths, may, alternatively, be used. Further, the diagrams presenting the spectroscopic data as illustrated herein have underwent a pre-processing step, which included at least one pre-processing step, preferably selected from baseline correction, total area correction, or standard normal variate (SNV) correction.Figure 7B shows a further diagram used for determining at least one selectivity ratio from the mean spectra of the Pa6 samples according to Figure 7A. In particular, Figure 7B schematically depicts a course of the signal over wavenumber as well as selected bands configured for discriminating Pa6 from at least one other component comprised in at least one particle of a particulate plastic-containing material Ms, wherein a respective band position, band height and band width (peak half-width) and, if appropriate, a border to an adjacent band, are indicated.Table 7 lists the selected bands for the Pa6 discrimination based on Figure 7B, wherein, for each selected band, the corresponding band position, a respective band range as indicated by a lower border and an upper border, and a corresponding selectivity are provided. The entries inTable 7 are sorted by their selectivity. As generally used, the term “selectivity” of a selected band relates to an area under the band as limited by the respective lower border and the upper border.Table 7Selected bands for Pa6 discriminationFigure 8A shows a diagram illustrating spectroscopic data as a course of a signal over wavenumber for mean spectra of polyamide 66 (Pa66) samples.Figure 8B shows a further diagram used for determining at least one selectivity ratio from the mean spectra of the Pa66 samples according to Figure 8A. In particular, Figure 8B schematically depicts a course of the signal over wavenumber as well as selected bands configured for discriminating Pa66 from at least one other component comprised in at least one particle of a particulate plastic-containing material Ms, wherein a respective band position, band height and band width (peak half-width) and, if appropriate, a border to an adjacent band, are indicated.Table 8 lists the selected bands for the Pa66 discrimination based on Figure 8B, wherein, for each selected band, the corresponding band position, a respective band range as indicated by a lower border and an upper border, and a corresponding selectivity are provided, wherein the entries in Table 8 are sorted by their selectivity.Table 8Selected bands for Pa66 discriminationFigure 9A shows a diagram illustrating spectroscopic data as a course of a signal over wavenumber for mean spectra of test samples comprising other polymers than PA6 and PA66.Figure 9B shows a further diagram used for determining at least one selectivity ratio from the mean spectra of the test samples comprising other polymers than PA6 and PA66 according to Figure 9A. In particular, Figure 9B schematically depicts a course of the signal over wavenumber as well as selected bands configured for discriminating the other polymers than PA6 and PA66 from the PA6 and PA66 comprised in at least one particle of a particulate plastic-containing material Ms, wherein a respective band position, band height and band width (peak half- width) and, if appropriate, a border to an adjacent band, are indicated.Table 9 lists the selected bands for the discrimination of the other polymers than PA6 and PA66 based on Figure 9B, wherein, for each selected band, the corresponding band position, a respective band range as indicated by a lower border and an upper border, and a corresponding selectivity are provided, wherein the entries in Table 9 are sorted by their selectivity.Table 9Selected bands for discrimination of other polymers than PA6 and PA66List of reference numbers110 overlapped mean spectra112 wavenumber114 spectroscopic data of the PA6 sample116 spectroscopic data of the PA66 sample118 ATR-FTIR spectroscopic data for the PA6 sample120 ATR-FTIR spectroscopic data for the PA66 sample122 variance importance in projection124 VIP for the PA6 sample126 VIP for the PA66 sample128 selectivity ratio130 selectivity ratio for the PA6 sample132 selectivity ratio for the PA66 sampleCited literatureEP 3 954473 A1E. H. Enlow et al. describe in “Discrimination of Nylon Polymers Using Attenuated Total Reflection Mid-infrared Spectra and Multivariate Statistical Techniques", Applied Spectroscopy, 2005;59(8):986-992M. N. Larsen et al. describe in ,, Classification of black plastic types by hyperspectral imaging based on long-wave infrared emission spectroscopy” , Polymer Testing, Volume 141 , 2024, 108629

Claims

Claims1 . A process for sorting a plastic-containing material, the process comprising(i) providing a particulate plastic-containing material Ms;(ii) subjecting the particulate material Ms according to (i) to sorting in a sorting unit Us, said sorting unit Us comprising(a) a classification section Sc for classifying particles of the particulate material Ms with respect to their polyamide 6 content and polyamide 66 content, said classification section Sc comprising(a.1) an optical system configured for measuring at least two wavelengths and determining spectroscopic data of the particles of the particulate material Ms therefrom;(a.2) a processing unit UP configured for discriminating polyamide 6 comprised in a particle of Ms from polyamide 66 comprised in said particle of Ms by performing a computer-implemented classification method using the spectroscopic data according to (a.1);(b) a sorting section Ss comprising a sorting device Ds configured for sorting particles of Ms classified in Sc according to their respective polyamide 6 and polyamide 66 content; wherein subjecting the particulate material Ms to sorting comprises(ii.1 ) feeding the particulate material Ms provided according to (i) to the classification section Sc according to (a);(11.2) determining spectroscopic data of x % of the particles of Ms with the optical system according to (a.1), comprising measuring at least two wavelengths of a given particle in a wavelength range Rw of from 0.9 to 12 pm, wherein x > 50;(11.3) classifying the particles spectroscopically examined according to (ii.2) in the processing UP according to (a.2), comprising performing the computer-implemented classification method using the spectroscopic data determined according to (ii.2), obtaining particles classified according to their polyamide 6 and polyamide 66 content;(11.4) sorting the particles classified according to (ii.3) according to their respective polyamide 6 and polyamide 66 content in Ss via the sorting device Ds, obtaining a sorted particulate plastic-containing material Ms.

2. The process of claim 1 , wherein the wavelength range Rw according to (ii.2) is of from 3 to 12 pm, preferably of from 4 to 12 pm, preferably of from 5 to 12 pm, more preferably of from 6 to 12 pm, more preferably of from 7 to 12 pm, more preferably of from 8 to 12 pm.

3. The process of claim 1 or 2, wherein the particulate plastic-containing material Ms comprises, preferably consists of a particulate plastic-containing waste material, more preferably of a particulate engineering plastic-containing waste material; wherein the particulate plastic-containing material Ms preferably comprises one or more of polyamide 6 and polyamide 66 and more preferably further comprises carbon black; wherein preferably from 50 to 100 weight-%, more preferably from 70 to 100 weight-%, more preferably from 80 to 100 weight-%, more preferably from 90 to 100 weight-%, more preferably from 95 to 100 weight-% of the particulate plastic-containing material Ms consists of one or more of polyamide 6 and polyamide 66 and preferably additionally carbon black, wherein, if less than 100 weight-% of the particulate plastic-containing material Ms consist of one or more of polyamide 6 and polyamide 66 and preferably additionally carbon black, the plastic-containing material Ms preferably further comprising one or more components selected from the group consisting of at least one polymer compound other than polyamide 6 and polyamide 66, wood, glass, and at least one metal, wherein more preferably, the at least one polymer compound comprises one or more of polypropylene, polycarbonate, polydimethylsiloxane, polyethylene, polyurethane, acrylonitrile-butadiene- styrene, acrylonitrile-butadiene-styrene, acrylonitrile styrene acrylate, polyethylene terephthalate, polyester, polystyrene, polyvinylchloride, a copolymer of two or more thereof including statistical copolymers, gradient copolymers, alternating copolymers, block copolymers, and graft copolymers, polymer blends of two or more thereof, a rubber material comprising one or more of a natural rubber material and a synthetic rubber material.

4. The process of any one of claims 1 to 3, preferably of claim 3, wherein the particulate material Ms provided according to (i) is obtainable or obtained from a method comprising(1.1) shredding a vehicle comprising polymeric vehicle parts, obtaining an automotive shredder residue (ASR);(1.2) preferably subjecting the automotive shredder residue according to (i.1) to one or more post-treatment steps (i.2.1) to (i.2.3):(i.2.1 ) separating metal fragments, including, but not restricted to one or more of ferrous metal fragments and non-ferrous metal fragments from the automotive shredder residue;(i.2.2.) separating the automotive shredder residue into a shredder light fraction and a shredder heavy fraction;(i.2.3) subjecting the automotive shredder residue, preferably one or more of the the shredder light fraction and the shredder heavy fraction according to(i.2.2) to an aqueous treatment, said aqueous treatment preferably comprising one or more of washing and aqueous density separation; wherein the particulate material Ms is obtained from (i.1), preferably from (i.2).

5. The process of any one of claims 1 to 4, wherein performing the computer-implemented classification method according to (ii.3) comprises(a) providing a computer-implemented trained classification model;(P) retrieving the trained model provided according to (a);(y) retrieving spectroscopic data of a particle of Ms determined according to (ii.2);(5) applying the trained model to the spectroscopic data according to (y), thereby discriminating polyamide 6 comprised from polyamide 66 and obtaining particles classified according to their polyamide 6 and polyamide 66 content; wherein providing the computer-implemented trained classification model according to (a) preferably comprises(a.1) providing a computer-implemented trainable model;(a.2) training the trainable model provided according to (a.1) with a computer-implemented training method, the training method comprising(a.2.1) retrieving a labelled spectroscopic training data set comprising spectroscopic data of polymer compounds determined by an optical system configured for measuring at least two wavelengths in the wavelength range Rw of polymer compounds having defined chemical composition;(a.2.2) training the trainable model on the labelled spectroscopic training data set, preferably by supervised training.

6. The process of claim 5, wherein the trainable model according to (a.1) comprises at least one model selected from the group consisting of a decision tree model, preferably at least one of an XGBoost model and a Random Forest model; a Support Vector Machine (SVM) model, preferably a linear kernel SVM; a nearest neighbors model, preferably a KNeigh- bors Classifier; a Bayes model, preferably a GaussianNB() model; a regression model, preferably at least one of a LogisticRegression, a linear and a nonlinear regression model including a regression model comprising transformed features which includes one or more of a log-transformed and polynomial model including a Partial Least Squares Discriminant Analysis (PLS-DA) model; an Artificial Neural Network (ANN), preferably a non-linear Artificial Neural Network (ANN), more preferably a deep learning architecture including one or more of Convolutional NN, Recurrent NN, a Long Short Term Memory NN; a kernel-based method; a tree regression model; a distributed gradient-boosting framework model, preferably a light gradient-boosting machine model (LightGBM classifier); a soft independentmodelling by class analogy (SIMCA); a Multi Curve Resolution (MCR) model; a t-distrib- uted stochastic neighbor embedding (t-SNE) model; a Uniform Manifold Approximation and Projection for Dimension Reduction (UMAP) model; a Classical Least Squares (CLS) model; a Simultaneous Component Analysis (ASCA-ANOVA) model; and a Multi-Level Simultaneous Component Analysis (MLSCA) model.

7. The process of claim 5 or 6, wherein the training method according to (a.2) further comprises pre-processing the spectroscopic data determined by CH to obtain the spectroscopic training data set, the pre-processing preferably comprising applying a Standard Normal Variate (SNV) to the spectroscopic data obtained, the training method preferably further comprising at least one cross validation in the spectroscopic training data set to obtain at least one pair of training data set and test data set and at least one validation step comprising testing the accuracy of the training by using the test data set.

8. The process of any one of claims 5 to 7, wherein applying the trained model according to (5) comprises attributing a classification probability to a given particle of Ms, wherein the particle is classified as a polyamide 6 particle if the polyamide 6 classification probability is at least 0.6, preferably at least 0.7, more preferably at least 0.8, more preferably at least 0.9, and wherein the particle is classified as a polyamide 66 particle if the polyamide 66 classification probability is at least 0.6, preferably at least 0.7, more preferably at least 0.8, more preferably at least 0.9.

9. The process of any one of claims 1 to 8, wherein the sorting device Ds comprised in the sorting section Ss comprises three outlet devices DOPA6, DOPA66 and DON, and wherein sorting the classified particles according to (ii.4) comprises controlling the sorting device Ds so that a particle classified as a polyamide 6 particle is passed to DOPA6, a particle classified as a polyamide 66 particle is passed to DOPA66, and a particle classified neither as a polyamide 6 particle nor as a polyamide 66 particle is passed to DON, wherein the sorting device Ds preferably comprises a first sorting subdevice Dssi, preferably a first pneumatic sorting subdevice or a first mechanical sorting subdevice, configured for actively removing a particle classified as a polyamide 6 particle to DOPA6 from the particulate material Ms or for actively removing a particle classified as a polyamide 66 particle to DOPA66 from the particulate material Ms; anda second sorting subdevice Dss2, preferably a second pneumatic sorting subdevice or a second mechanical sorting subdevice, configured for actively removing a particle classified as a polyamide 66 particle to DOPA66 or for actively removing a particle classified as a polyamide 6 particle to DOPA6 from the remainder material; wherein, if the particulate material Ms provided according to (i) comprises particles classified as polyamide 6 particles according to (ii.3), particles classified as polyamide 66 particles according to (ii.3), and optionally particles classified neither as polyamide 6 particles nor as polyamide 66 particles, (ii.4) preferably comprises(11.4.1) actively removing particles classified as polyamide 6 particles to DOPA6 from the particulate material Ms via Dssi, obtaining sorted-out particles classified as polyamide 6 particles and further obtaining a first remainder material MSRI comprising particles classified as polyamide 66 particles and optionally particles neither classified as polyamide 66 particles nor classified as polyamide 6 particles;(11.4.2) actively removing particles classified as polyamide 66 particles to DOPA66 from the remained material according to (ii.4.1) via Dss2, obtaining sorted-out particles classified as polyamide 66 particles and optionally further obtaining a second remainder material MSR2 comprising particles classified neither as polyamide 66 particles nor as polyamide 6 particles, said second remainder stream being passed to DON; or(ii.4.1’) actively removing particles classified as polyamide 66 particles to DOPA66 from the particulate material Ms via Dssi, obtaining sorted-out particles classified as polyamide 66 particles and further obtaining a first remainder material MSRI comprising particles classified as polyamide 6 particles and optionally particles neither classified as polyamide 66 particles nor classified as polyamide 6 particles;(ii.4.2’) actively removing particles classified as polyamide 6 particles to DOPA6 from the remained material according to (ii.4.1) via Dss2, obtaining sorted-out particles classified as polyamide 6 particles and optionally further obtaining a second remainder material MSR2 comprising particles classified neither as polyamide 6 particles nor as polyamide 66 particles, said second remainder stream being passed to DON; wherein controlling Ds, and preferably Dssi and Dss2, is preferably computer-controlled.

10. Sorted plastic-containing material, preferably sorted plastic-containing waste material, obtainable or obtained by a process according to any one of claims 1 to 9.

11. A computer-implemented classification method for discriminating polyamide 6 from polyamide 66, preferably for discriminating polyamide 6 comprised in a particle of a particulate plastic-containing material Ms from polyamide 66 comprised in said particle of Ms, said computer-implemented classification method preferably being carried out in step (ii.3) of the sorting process according to any one of claims 1 to 9, the method comprising(a) providing a computer-implemented trained classification model;(P) retrieving the trained model provided according to (a);(y) retrieving spectroscopic data of polymer compounds comprising one or more of polyamide 6 and polyamide 66, said spectroscopic data being determined by an optical system configured for measuring at least two wavelengths in a wavelength range Rw of from 0.9 to 12 pm;(5) applying the trained model to the spectroscopic data according to (y), thereby discriminating polyamide 6 from polyamide 66; wherein Rw according to (y) is preferably of from 3 to 12 pm, more preferably of from 4 to 12 pm, more preferably of from 5 to 12 pm, more preferably of from 6 to 12 pm, more preferably of from 7 to 12 pm, more preferably of from 8 to 12 pm; wherein providing the trained model trained by using at least one computer-implemented training method according to (a) preferably comprises (a.1) providing a computer-implemented trainable model;(a.2) training the trainable model provided according to (a.1) with a computer-implemented training method, the training method comprising(a.2.1) retrieving a labelled spectroscopic training data set, the spectroscopic training data set comprising spectroscopic data of polymer compounds determined by an optical system configured for measuring at least two wavelengths in the wavelength range Rw of polymer compounds having defined chemical composition;(a.2.2) training the trainable model on the labelled spectroscopic training data set, preferably by supervised training.

12. A method for providing a trained model for discriminating polyamide 6 from polyamide 66, preferably for discriminating polyamide 6 comprised a particle of a particulate plastic-containing material Ms from polyamide 66 comprised in said particle of Ms, said method preferably being carried out as step (a) according to claim 11 , the method comprising(a.1) providing a computer-implemented trainable model;(a.2) training the trainable model provided according to (a.1) with a computer-implemented training method, the training method comprising(a.2.1) retrieving a labelled spectroscopic training data set, the spectroscopic training data set comprising spectroscopic data of polymer compounds determined by an optical system configured for measuring at least two wavelengths in a wavelength range Rw of from 0.9 to 12 pm of polymer compounds having defined chemical composition;(a.2.2) training the trainable model on the labelled spectroscopic training data set, preferably by supervised training; wherein Rw according to (a.2.1) is preferably of from 3 to 12 pm, more preferably of from 4 to 12 pm, more preferably of from 5 to 12 pm, more preferably of from 6 to 12 pm, more preferably of from 7 to 12 pm, more preferably of from 8 to 12 pm.

13. A sorting unit Us for carrying out a sorting process according to any one of claims 1 to 9, the sorting unit Us comprising(a) a classification section Sc for classifying particles of the particulate material Ms with respect to their polyamide 6 content and polyamide 66 content, said classification section Sc comprising(a.1) an optical system configured for measuring at least two wavelengths and determining spectroscopic data of the particles of the sorting material therefrom, said determining spectroscopic data of the particles of Ms with the optical system comprising measuring at least two wavelengths of a given particle in a wavelength range Rw of from 0.9 to 12 pm;(a.2) a processing unit UP configured for discriminating polyamide 6 comprised a particle of Ms from polyamide 66 comprised in said particle of Ms by performing a computer-implemented classification method using the spectroscopic data according to (a.1); wherein Rw according to (a.1) is preferably of from 3 to 12 pm, more preferably of from 4 to 12 pm, more preferably of from 5 to 12 pm, more preferably of from 6 to 12 pm, more preferably of from 7 to 12 pm, more preferably of from 8 to 12 pm.(b) a sorting section Ss comprising a sorting device Ds configured for sorting particles of Ms classified in Sc according to their respective polyamide 6 and polyamide 66 content, the sorting section Ss preferably further comprising three outlet devices DOPA6, DQPA66 and DON; wherein the sorting device Ds according to (b) preferably comprises a first sorting subdevice Dssi, preferably a first pneumatic sorting subdevice or a first mechanical sorting subdevice, configured for actively removing a particle classifiedas a polyamide 6 particle to DOPA6 from the particulate material Ms or for actively removing a particle classified as a polyamide 66 particle to DOPA66 from the particulate material Ms; and a second sorting subdevice Dss2, preferably a second pneumatic sorting subdevice or a second mechanical sorting subdevice, configured for actively removing a particle classified as a polyamide 66 particle to DOPA66 or for actively removing a particle classified as a polyamide 6 particle to DOPA6 from the remainder material; wherein controlling Ds is preferably computer-controlled.

14. A method for recycling plastic-containing material, comprising providing sorted a plasticcontaining material, preferably sorted plastic-containing waste material, said sorted material being obtainable or obtained by a process according to any one of claims 1 to 9 and / or by using the sorting unit according to claim 13, and recycling said sorted plastic-containing material to obtain a polymer; wherein the method for recycling preferably comprises, more preferably consists of one or more of mechanical recycling, chemical recycling and sol- vent-based recycling.

15. A process, preferably according to any one of claims 1 to 9, comprising the step of converting a chemical material obtainable or obtained by the process according to any one of claims 1 to 9 to obtain a product Q, and / or a process, comprising the step of using the sorting unit according to claim 13 to obtain a chemical material; and preferably converting the chemical material to obtain a product Q.

Images